In [1]:
%run NADINE_classification_sea.ipynb
Number of input: 3 Number of output: 2 Number of batch: 100 All Data
100% (100 of 100) |######################| Elapsed Time: 0:03:17 ETA: 00:00:00
=== Performance result === Accuracy: 92.04646464646466 (+/-) 7.245762441583441 Testing Loss: 0.24809447032484142 (+/-) 0.17327467956558293 Precision: 0.9209290579001399 Recall: 0.9204646464646464 F1 score: 0.9196410650965958 Testing Time: 0.002747598320546776 (+/-) 0.004207666574854691 Training Time: 1.9877013413593023 (+/-) 0.03938866113608207 === Average network evolution === Total hidden node: 9.080808080808081 (+/-) 2.5412824555178 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 13 No. of parameters : 52 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=2, bias=True) ) No. of inputs : 13 No. of output : 2 No. of parameters : 28 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:03:16 ETA: 00:00:00
=== Performance result === Accuracy: 91.59494949494949 (+/-) 7.566611583531633 Testing Loss: 0.2566300773560399 (+/-) 0.17824953720371742 Precision: 0.9166665640997699 Recall: 0.915949494949495 F1 score: 0.914941185008031 Testing Time: 0.0026094070588699495 (+/-) 0.0008937288137483986 Training Time: 1.9765822381684275 (+/-) 0.0707110478646652 === Average network evolution === Total hidden node: 10.818181818181818 (+/-) 2.0068752351270844 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 14 No. of parameters : 56 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=2, bias=True) ) No. of inputs : 14 No. of output : 2 No. of parameters : 30 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:03:18 ETA: 00:00:00
=== Performance result === Accuracy: 91.86060606060607 (+/-) 7.252903538149439 Testing Loss: 0.25204279887104275 (+/-) 0.17577178453303247 Precision: 0.9190163401241277 Recall: 0.9186060606060606 F1 score: 0.9177674044007206 Testing Time: 0.0025177387276081125 (+/-) 0.0007885767857022272 Training Time: 1.9954934240591646 (+/-) 0.05999641203153533 === Average network evolution === Total hidden node: 9.292929292929292 (+/-) 2.1379331341063037 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 12 No. of parameters : 48 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:03:13 ETA: 00:00:00
=== Performance result === Accuracy: 92.0050505050505 (+/-) 7.388461844781541 Testing Loss: 0.24668141902245658 (+/-) 0.17388963722523088 Precision: 0.9206835395137225 Recall: 0.920050505050505 F1 score: 0.9191573991246542 Testing Time: 0.002729856606685754 (+/-) 0.004720559681870977 Training Time: 1.953931415923918 (+/-) 0.03738359701839315 === Average network evolution === Total hidden node: 5.525252525252525 (+/-) 1.6957667694087522 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 8 No. of parameters : 32 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:03:16 ETA: 00:00:00
=== Performance result === Accuracy: 91.7111111111111 (+/-) 7.244125295070801 Testing Loss: 0.2522579489106482 (+/-) 0.17495033080632222 Precision: 0.9177422975813482 Recall: 0.9171111111111111 F1 score: 0.9161582395155624 Testing Time: 0.0025168886088361643 (+/-) 0.0006976634724533672 Training Time: 1.9813676434333878 (+/-) 0.08373462430746167 === Average network evolution === Total hidden node: 10.171717171717171 (+/-) 2.0550032762656874 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 13 No. of parameters : 52 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=2, bias=True) ) No. of inputs : 13 No. of output : 2 No. of parameters : 28 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 92.13183673469389 Std Accuracy: 6.800135554912209 Hidden Node mean 9.00204081632653 Hidden Node std: 2.7888253540651995 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (100 of 100) |######################| Elapsed Time: 0:01:33 ETA: 00:00:00
=== Performance result === Accuracy: 90.84848484848483 (+/-) 8.553342587637163 Testing Loss: 0.26605809062267793 (+/-) 0.17459025784965898 Precision: 0.9097661796436708 Recall: 0.9084848484848485 F1 score: 0.9071095578262883 Testing Time: 0.0029125526697948725 (+/-) 0.005091869841401095 Training Time: 0.934997760888302 (+/-) 0.09522105697186267 === Average network evolution === Total hidden node: 9.0 (+/-) 1.964328347956588 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 12 No. of parameters : 48 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:01:34 ETA: 00:00:00
=== Performance result === Accuracy: 91.20505050505052 (+/-) 7.760693160458192 Testing Loss: 0.2693159354831835 (+/-) 0.17317394021816723 Precision: 0.9125088757185745 Recall: 0.912050505050505 F1 score: 0.9110555739384203 Testing Time: 0.0029651468450372868 (+/-) 0.004600721437592303 Training Time: 0.9536460914997139 (+/-) 0.08014681759194285 === Average network evolution === Total hidden node: 8.878787878787879 (+/-) 1.9083692719798533 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 12 No. of parameters : 48 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:01:40 ETA: 00:00:00
=== Performance result === Accuracy: 90.88686868686868 (+/-) 8.760282806013896 Testing Loss: 0.268101196472693 (+/-) 0.1766333648120577 Precision: 0.9097953334219658 Recall: 0.9088686868686868 F1 score: 0.907621646110993 Testing Time: 0.0022969101414535985 (+/-) 0.000856577416973236 Training Time: 1.0072671789111514 (+/-) 0.01612809927573245 === Average network evolution === Total hidden node: 5.565656565656566 (+/-) 1.7242279265395608 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 8 No. of parameters : 32 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:01:39 ETA: 00:00:00
=== Performance result === Accuracy: 91.02727272727276 (+/-) 8.235618848302094 Testing Loss: 0.2678614988019972 (+/-) 0.17114389009308614 Precision: 0.9110367155350154 Recall: 0.9102727272727272 F1 score: 0.909118276531031 Testing Time: 0.0028200558941773694 (+/-) 0.004830555760948444 Training Time: 0.9993915028042264 (+/-) 0.012765175556011258 === Average network evolution === Total hidden node: 4.696969696969697 (+/-) 1.5403957608653134 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 7 No. of parameters : 28 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 90.03333333333333 (+/-) 9.84078300243482 Testing Loss: 0.28078381371016453 (+/-) 0.190405711632873 Precision: 0.9025464377352657 Recall: 0.9003333333333333 F1 score: 0.8984366335839261 Testing Time: 0.0023282802466190224 (+/-) 0.0007924938211500075 Training Time: 0.9931801160176595 (+/-) 0.01360737271221153 === Average network evolution === Total hidden node: 5.747474747474747 (+/-) 2.3925111880022305 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 10 No. of parameters : 40 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=2, bias=True) ) No. of inputs : 10 No. of output : 2 No. of parameters : 22 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 91.07775510204083 Std Accuracy: 8.26208793231163 Hidden Node mean 6.795918367346939 Hidden Node std: 2.6432786678262072 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (100 of 100) |######################| Elapsed Time: 0:00:51 ETA: 00:00:00
=== Performance result === Accuracy: 89.24747474747475 (+/-) 9.529140245852187 Testing Loss: 0.3055352962227783 (+/-) 0.16683987778905285 Precision: 0.8953348558420224 Recall: 0.8924747474747474 F1 score: 0.8901141465922442 Testing Time: 0.0024453654433741713 (+/-) 0.000899871271626218 Training Time: 0.5140980614556206 (+/-) 0.012067787521766522 === Average network evolution === Total hidden node: 7.434343434343434 (+/-) 1.7417140991749975 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 10 No. of parameters : 40 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=2, bias=True) ) No. of inputs : 10 No. of output : 2 No. of parameters : 22 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:00:52 ETA: 00:00:00
=== Performance result === Accuracy: 88.13535353535357 (+/-) 10.856793368612582 Testing Loss: 0.3240241948703323 (+/-) 0.17700968466881314 Precision: 0.8866859549087837 Recall: 0.8813535353535353 F1 score: 0.8778278949417561 Testing Time: 0.0025330577233825067 (+/-) 0.0047486244661193635 Training Time: 0.5220378601189816 (+/-) 0.021475664792920037 === Average network evolution === Total hidden node: 2.7777777777777777 (+/-) 1.0876306892846717 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 5 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:00:51 ETA: 00:00:00
=== Performance result === Accuracy: 88.9040404040404 (+/-) 10.413481153569217 Testing Loss: 0.3052358489596482 (+/-) 0.17532217613379503 Precision: 0.8922409791889865 Recall: 0.8890404040404041 F1 score: 0.8864468519580697 Testing Time: 0.0023789622566916728 (+/-) 0.0008365835694093035 Training Time: 0.5202701766081531 (+/-) 0.019487244057278474 === Average network evolution === Total hidden node: 6.737373737373737 (+/-) 1.6671869405270094 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 9 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:00:51 ETA: 00:00:00
=== Performance result === Accuracy: 87.13939393939395 (+/-) 12.324893247898112 Testing Loss: 0.3274126123599332 (+/-) 0.19862582560956255 Precision: 0.8796176920553247 Recall: 0.8713939393939394 F1 score: 0.8665672992318848 Testing Time: 0.0020268517311173256 (+/-) 0.0008924830360491278 Training Time: 0.5135503898967396 (+/-) 0.01374306886910887 === Average network evolution === Total hidden node: 3.1717171717171717 (+/-) 1.4567183545396163 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 6 No. of parameters : 24 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (100 of 100) |######################| Elapsed Time: 0:00:51 ETA: 00:00:00
=== Performance result === Accuracy: 90.77777777777777 (+/-) 7.828871141138848 Testing Loss: 0.28165094990922945 (+/-) 0.16283372396354528 Precision: 0.9092196717509192 Recall: 0.9077777777777778 F1 score: 0.9063332375599721 Testing Time: 0.0031206294743701666 (+/-) 0.004686521857226626 Training Time: 0.5138730954642248 (+/-) 0.018132758133208943 === Average network evolution === Total hidden node: 7.929292929292929 (+/-) 1.5907086614165276 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=11, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 11 No. of parameters : 44 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=11, out_features=2, bias=True) ) No. of inputs : 11 No. of output : 2 No. of parameters : 24 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 89.08469387755102 Std Accuracy: 10.126261594778247 Hidden Node mean 5.6204081632653065 Hidden Node std: 2.676197393030636 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
89% (89 of 100) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 67.47474747474747 (+/-) 7.314885106968543 Testing Loss: 0.5595584740542402 (+/-) 0.04018779356611432 Precision: 0.7680273804304095 Recall: 0.6747474747474748 F1 score: 0.583256426337157 Testing Time: 0.00295170389040552 (+/-) 0.0018365735723885267 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 5 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
99% (99 of 100) |###################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 67.3010101010101 (+/-) 7.308380341917092 Testing Loss: 0.5659842734987085 (+/-) 0.04105067714654622 Precision: 0.7670244652694179 Recall: 0.6730101010101011 F1 score: 0.5798968120305256 Testing Time: 0.0020083176969277737 (+/-) 0.005079080709222568 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 3 No. of parameters : 12 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
92% (92 of 100) |##################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 63.347474747474756 (+/-) 7.370091668280475 Testing Loss: 0.6104204203143264 (+/-) 0.024466945101211213 Precision: 0.7489237630283784 Recall: 0.6334747474747475 F1 score: 0.4948721992567769 Testing Time: 0.00159186546248619 (+/-) 0.0007225927284669849 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 6.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 6 No. of parameters : 24 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
82% (82 of 100) |################## | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 65.50505050505052 (+/-) 7.413301909370037 Testing Loss: 0.6115767817304592 (+/-) 0.025476510849458286 Precision: 0.7591664775237807 Recall: 0.6550505050505051 F1 score: 0.5431859017795544 Testing Time: 0.0018591447310014205 (+/-) 0.0007259296935056934 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 8 No. of parameters : 32 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
90% (90 of 100) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 65.53636363636365 (+/-) 7.3618379148097395 Testing Loss: 0.5628098797316503 (+/-) 0.049897174156126864 Precision: 0.7586683659827788 Recall: 0.6553636363636364 F1 score: 0.5439290938022063 Testing Time: 0.002275642722544044 (+/-) 0.00475496788604065 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 7 No. of parameters : 28 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 65.82306122448979 Std Accuracy: 7.540065691285761 Hidden Node mean 5.8 Hidden Node std: 1.7204650534085253 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [2]:
%run NADINE_classification_hyperplane.ipynb
Number of input: 4 Number of output: 2 Number of batch: 120 All Data
100% (120 of 120) |######################| Elapsed Time: 0:03:32 ETA: 00:00:00
=== Performance result === Accuracy: 92.15042016806723 (+/-) 2.8990254505530983 Testing Loss: 0.29649631203222676 (+/-) 0.05067234374492242 Precision: 0.9215042597324717 Recall: 0.9215042016806723 F1 score: 0.9215042160650396 Testing Time: 0.002300685193358349 (+/-) 0.000705926809147076 Training Time: 1.7797974277945126 (+/-) 0.03483840790539562 === Average network evolution === Total hidden node: 4.722689075630252 (+/-) 0.5639651003550427 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:03:31 ETA: 00:00:00
=== Performance result === Accuracy: 92.38655462184875 (+/-) 4.0025816116527615 Testing Loss: 0.28320085351206675 (+/-) 0.06635055100422449 Precision: 0.923880336030871 Recall: 0.9238655462184874 F1 score: 0.9238644855502037 Testing Time: 0.0020635949463403526 (+/-) 0.0008180129898153774 Training Time: 1.773071429308723 (+/-) 0.03246699468754135 === Average network evolution === Total hidden node: 2.689075630252101 (+/-) 0.4628719110561482 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:03:30 ETA: 00:00:00
=== Performance result === Accuracy: 91.92521008403362 (+/-) 6.052448889574097 Testing Loss: 0.2916631070004792 (+/-) 0.07360288900962149 Precision: 0.9192630322025698 Recall: 0.9192521008403362 F1 score: 0.919251206655974 Testing Time: 0.0023710387093680246 (+/-) 0.004508675639463742 Training Time: 1.767293781793418 (+/-) 0.022291765765204013 === Average network evolution === Total hidden node: 2.6218487394957983 (+/-) 0.5019557822880026 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:03:30 ETA: 00:00:00
=== Performance result === Accuracy: 91.5747899159664 (+/-) 7.0038071098149866 Testing Loss: 0.2941112954075597 (+/-) 0.09029855397836757 Precision: 0.9158714242448224 Recall: 0.9157478991596638 F1 score: 0.9157428938392136 Testing Time: 0.001949346366048861 (+/-) 0.0006170568800573533 Training Time: 1.7663562518207967 (+/-) 0.02545448962813301 === Average network evolution === Total hidden node: 2.6470588235294117 (+/-) 0.47788461203740956 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:03:31 ETA: 00:00:00
=== Performance result === Accuracy: 92.61680672268908 (+/-) 2.537063720916055 Testing Loss: 0.27964958961771313 (+/-) 0.0572829309409839 Precision: 0.926193076707485 Recall: 0.9261680672268907 F1 score: 0.9261664898737734 Testing Time: 0.0020519064254119618 (+/-) 0.0006537969101582346 Training Time: 1.7728120339016955 (+/-) 0.02470586330893761 === Average network evolution === Total hidden node: 2.563025210084034 (+/-) 0.4960119180965951 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 92.36474576271186 Std Accuracy: 3.989555582884229 Hidden Node mean 3.047457627118644 Hidden Node std: 0.9782994467577771 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (120 of 120) |######################| Elapsed Time: 0:01:47 ETA: 00:00:00
=== Performance result === Accuracy: 90.95042016806723 (+/-) 5.672799228243759 Testing Loss: 0.3199200607648417 (+/-) 0.08689299819893508 Precision: 0.9097760950749174 Recall: 0.9095042016806723 F1 score: 0.9094871294846018 Testing Time: 0.0022505591897403494 (+/-) 0.004076026827974053 Training Time: 0.8989008494785854 (+/-) 0.009925686023113937 === Average network evolution === Total hidden node: 2.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=2, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 2 No. of parameters : 10 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=2, out_features=2, bias=True) ) No. of inputs : 2 No. of output : 2 No. of parameters : 6 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:01:47 ETA: 00:00:00
=== Performance result === Accuracy: 90.75294117647057 (+/-) 6.817597392529412 Testing Loss: 0.3187672264185272 (+/-) 0.08624936502083748 Precision: 0.9076082111257892 Recall: 0.9075294117647059 F1 score: 0.9075237969767198 Testing Time: 0.0019830775861980534 (+/-) 0.0006825015700865035 Training Time: 0.8996561655477315 (+/-) 0.01522871386239103 === Average network evolution === Total hidden node: 2.6554621848739495 (+/-) 0.5566552995522419 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:01:47 ETA: 00:00:00
=== Performance result === Accuracy: 88.30504201680674 (+/-) 10.15210247496055 Testing Loss: 0.36226710476795165 (+/-) 0.12782271646259816 Precision: 0.883458371358108 Recall: 0.8830504201680672 F1 score: 0.8830158091366551 Testing Time: 0.00185481840822877 (+/-) 0.0006891938453977504 Training Time: 0.901871108207382 (+/-) 0.016199070713873557 === Average network evolution === Total hidden node: 2.1260504201680672 (+/-) 0.3319061791282604 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=2, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 2 No. of parameters : 10 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=2, out_features=2, bias=True) ) No. of inputs : 2 No. of output : 2 No. of parameters : 6 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:01:47 ETA: 00:00:00
=== Performance result === Accuracy: 91.19915966386552 (+/-) 5.392049793400228 Testing Loss: 0.3135913302417563 (+/-) 0.07213994704383872 Precision: 0.912071960078489 Recall: 0.9119915966386555 F1 score: 0.9119862179790043 Testing Time: 0.002477130970033277 (+/-) 0.004041488405764773 Training Time: 0.8993637742114668 (+/-) 0.014733840329427744 === Average network evolution === Total hidden node: 3.7394957983193278 (+/-) 0.43890974309917974 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 4 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:01:47 ETA: 00:00:00
=== Performance result === Accuracy: 90.49747899159665 (+/-) 6.615190847363184 Testing Loss: 0.31880455658215434 (+/-) 0.07534365667646682 Precision: 0.90501207833305 Recall: 0.9049747899159664 F1 score: 0.9049733920030214 Testing Time: 0.002207309258084337 (+/-) 0.0006200870986661813 Training Time: 0.8995499450619481 (+/-) 0.010604531889417547 === Average network evolution === Total hidden node: 4.1344537815126055 (+/-) 0.38728465955640046 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 4 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 90.64813559322035 Std Accuracy: 6.408370181910073 Hidden Node mean 2.9338983050847456 Hidden Node std: 0.9301193962890245 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (120 of 120) |######################| Elapsed Time: 0:00:55 ETA: 00:00:00
=== Performance result === Accuracy: 88.26554621848742 (+/-) 9.106027526132754 Testing Loss: 0.36755985109245076 (+/-) 0.11863846983396353 Precision: 0.8827062529273908 Recall: 0.8826554621848739 F1 score: 0.8826503110354769 Testing Time: 0.0020083619766876476 (+/-) 0.0007108639164210289 Training Time: 0.46482718291402864 (+/-) 0.011835648885369858 === Average network evolution === Total hidden node: 2.3361344537815127 (+/-) 0.8723320519234269 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=2, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 2 No. of parameters : 10 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=2, out_features=2, bias=True) ) No. of inputs : 2 No. of output : 2 No. of parameters : 6 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:00:55 ETA: 00:00:00
=== Performance result === Accuracy: 89.3470588235294 (+/-) 6.665398215132213 Testing Loss: 0.33770105092465375 (+/-) 0.08491642032580225 Precision: 0.8934785696167269 Recall: 0.8934705882352941 F1 score: 0.8934695968065305 Testing Time: 0.002765956045198841 (+/-) 0.0041384509030663345 Training Time: 0.46240100740384654 (+/-) 0.011405785031031476 === Average network evolution === Total hidden node: 5.621848739495798 (+/-) 0.4849256486135633 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 6 No. of parameters : 30 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:00:55 ETA: 00:00:00
=== Performance result === Accuracy: 89.27815126050417 (+/-) 6.691423830896029 Testing Loss: 0.33938866002219065 (+/-) 0.08444331092705229 Precision: 0.8927837368943965 Recall: 0.892781512605042 F1 score: 0.8927811100686223 Testing Time: 0.002402982791932691 (+/-) 0.000732108476588942 Training Time: 0.4610254984943807 (+/-) 0.007079709366170477 === Average network evolution === Total hidden node: 5.630252100840337 (+/-) 0.48273635685193506 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:00:55 ETA: 00:00:00
=== Performance result === Accuracy: 90.61344537815125 (+/-) 4.542847816142414 Testing Loss: 0.3321104891159955 (+/-) 0.06812160804703489 Precision: 0.9061632536475084 Recall: 0.9061344537815126 F1 score: 0.9061320780611632 Testing Time: 0.0026450838361467633 (+/-) 0.0007219116338995319 Training Time: 0.4614076754626106 (+/-) 0.007348057996985408 === Average network evolution === Total hidden node: 8.504201680672269 (+/-) 0.49998234556784926 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 9 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (120 of 120) |######################| Elapsed Time: 0:00:55 ETA: 00:00:00
=== Performance result === Accuracy: 86.9529411764706 (+/-) 9.02386959357891 Testing Loss: 0.3890424458419575 (+/-) 0.11649172311489467 Precision: 0.8697079012034705 Recall: 0.8695294117647059 F1 score: 0.8695109646439012 Testing Time: 0.0023615861139377626 (+/-) 0.0041331417854312625 Training Time: 0.46273432058446545 (+/-) 0.010790581342372464 === Average network evolution === Total hidden node: 3.4873949579831933 (+/-) 0.960265598924995 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=2, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 2 No. of parameters : 10 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=2, out_features=2, bias=True) ) No. of inputs : 2 No. of output : 2 No. of parameters : 6 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 89.16237288135594 Std Accuracy: 6.9281304101181265 Hidden Node mean 5.110169491525424 Hidden Node std: 2.231454258295662 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
97% (117 of 120) |##################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 65.37899159663866 (+/-) 3.6903661531706033 Testing Loss: 0.6363926474787608 (+/-) 0.017221841824245285 Precision: 0.7111893381651403 Recall: 0.6537899159663866 F1 score: 0.6282848206557048 Testing Time: 0.0018342402802795923 (+/-) 0.0008572280771557304 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 7 No. of parameters : 35 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
87% (105 of 120) |################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 71.37563025210083 (+/-) 1.5731981228932126 Testing Loss: 0.6305836579378914 (+/-) 0.003307096053899442 Precision: 0.7945577330524731 Recall: 0.7137563025210084 F1 score: 0.6925057489860829 Testing Time: 0.0017085796644707688 (+/-) 0.0007305651249097896 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
86% (104 of 120) |################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 50.035294117647055 (+/-) 4.360278848443511 Testing Loss: 0.6929996649758154 (+/-) 0.005599663744895442 Precision: 0.5004176250587449 Recall: 0.5003529411764706 F1 score: 0.4996094878463812 Testing Time: 0.002037116459437779 (+/-) 0.004819470633333363 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 4 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
97% (117 of 120) |##################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 77.30504201680674 (+/-) 2.2894294448854584 Testing Loss: 0.6188914159766766 (+/-) 0.005273408751214823 Precision: 0.7774206088596947 Recall: 0.7730504201680672 F1 score: 0.7721846291260479 Testing Time: 0.0014826049323843308 (+/-) 0.000694594772147449 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
99% (119 of 120) |##################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 49.903361344537814 (+/-) 1.6931861851743946 Testing Loss: 0.6965975070200047 (+/-) 0.00217992601564544 Precision: 0.28527484247922935 Recall: 0.49903361344537817 F1 score: 0.3336821284924525 Testing Time: 0.0016177261576933019 (+/-) 0.0006838670486240806 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 4 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 62.80593220338983 Std Accuracy: 11.519702299979345 Hidden Node mean 4.6 Hidden Node std: 1.3564659966250538 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [3]:
%run NADINE_classification_weather.ipynb
Number of input: 8 Number of output: 2 Number of batch: 18 All Data
100% (18 of 18) |########################| Elapsed Time: 0:00:30 ETA: 00:00:00
=== Performance result === Accuracy: 71.32941176470588 (+/-) 3.556727627846294 Testing Loss: 0.5403722156496609 (+/-) 0.04329136760952233 Precision: 0.6913702849107582 Recall: 0.7132941176470589 F1 score: 0.6890845092093629 Testing Time: 0.0016904297996969784 (+/-) 0.0005617839532356606 Training Time: 1.77819842450759 (+/-) 0.030996939894026028 === Average network evolution === Total hidden node: 7.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 8 No. of parameters : 72 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:30 ETA: 00:00:00
=== Performance result === Accuracy: 71.67058823529412 (+/-) 3.1112253507888172 Testing Loss: 0.5406189823851866 (+/-) 0.04603602344216499 Precision: 0.6946387747005545 Recall: 0.7167058823529412 F1 score: 0.6865026228408925 Testing Time: 0.00197604123283835 (+/-) 0.000591063307092828 Training Time: 1.7654989466947668 (+/-) 0.009852571421042795 === Average network evolution === Total hidden node: 6.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:30 ETA: 00:00:00
=== Performance result === Accuracy: 71.09411764705884 (+/-) 2.462774759406743 Testing Loss: 0.5428671065498801 (+/-) 0.04030093529778865 Precision: 0.6884534709358736 Recall: 0.7109411764705882 F1 score: 0.6604607546103711 Testing Time: 0.0019231824313893037 (+/-) 0.00040929424898422423 Training Time: 1.7815885543823242 (+/-) 0.01884339151193584 === Average network evolution === Total hidden node: 6.9411764705882355 (+/-) 0.4159451654038515 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 8 No. of parameters : 72 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:30 ETA: 00:00:00
=== Performance result === Accuracy: 71.70588235294119 (+/-) 3.48027561251984 Testing Loss: 0.5454000900773441 (+/-) 0.040067734513748746 Precision: 0.6952006889722699 Recall: 0.7170588235294117 F1 score: 0.6816144853217456 Testing Time: 0.0018056280472699333 (+/-) 0.00037103232072078693 Training Time: 1.7704437059514664 (+/-) 0.029176315804728674 === Average network evolution === Total hidden node: 4.352941176470588 (+/-) 0.47788461203740945 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:30 ETA: 00:00:00
=== Performance result === Accuracy: 70.69411764705882 (+/-) 2.981999050821505 Testing Loss: 0.5511516472872566 (+/-) 0.038062386796295644 Precision: 0.6819316781751159 Recall: 0.7069411764705882 F1 score: 0.6780131122610347 Testing Time: 0.0019874993492575255 (+/-) 0.00034309252911043936 Training Time: 1.7652401503394632 (+/-) 0.011489047161253723 === Average network evolution === Total hidden node: 7.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 8 No. of parameters : 72 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 71.21124999999999 Std Accuracy: 3.243223464009226 Hidden Node mean 6.3125 Hidden Node std: 1.0907995920424614 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (18 of 18) |########################| Elapsed Time: 0:00:15 ETA: 00:00:00
=== Performance result === Accuracy: 69.6058823529412 (+/-) 3.5470734562179893 Testing Loss: 0.5742993828128365 (+/-) 0.04396989162949954 Precision: 0.6640885048900774 Recall: 0.6960588235294117 F1 score: 0.6565126766136917 Testing Time: 0.0016372484319350298 (+/-) 0.00047316968865892476 Training Time: 0.8943946221295525 (+/-) 0.006913907754488262 === Average network evolution === Total hidden node: 5.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:15 ETA: 00:00:00
=== Performance result === Accuracy: 70.37647058823529 (+/-) 2.482297183650354 Testing Loss: 0.5670202146558201 (+/-) 0.029485407771790394 Precision: 0.6777116478403404 Recall: 0.703764705882353 F1 score: 0.6429458355402672 Testing Time: 0.001925510518691119 (+/-) 0.0005383760153305175 Training Time: 0.8952875978806439 (+/-) 0.007619511279858918 === Average network evolution === Total hidden node: 5.705882352941177 (+/-) 0.5703152773430975 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:15 ETA: 00:00:00
=== Performance result === Accuracy: 70.51764705882353 (+/-) 2.3746789985218557 Testing Loss: 0.55905381546301 (+/-) 0.029895203191934623 Precision: 0.679285954206223 Recall: 0.7051764705882353 F1 score: 0.6485433007137573 Testing Time: 0.001978032729204963 (+/-) 0.000341632885297832 Training Time: 0.9023997783660889 (+/-) 0.012391396872824055 === Average network evolution === Total hidden node: 6.529411764705882 (+/-) 0.4991341984846218 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:15 ETA: 00:00:00
=== Performance result === Accuracy: 70.17058823529412 (+/-) 3.13908280082731 Testing Loss: 0.5744826109970317 (+/-) 0.03653259881557109 Precision: 0.6741783808719745 Recall: 0.7017058823529412 F1 score: 0.6379157760150654 Testing Time: 0.0017485618591308594 (+/-) 0.000640537133346919 Training Time: 0.8928547746994916 (+/-) 0.003885215141072362 === Average network evolution === Total hidden node: 6.647058823529412 (+/-) 0.47788461203740945 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:15 ETA: 00:00:00
=== Performance result === Accuracy: 69.72941176470589 (+/-) 3.123302307020257 Testing Loss: 0.5642991960048676 (+/-) 0.0296759048046517 Precision: 0.6671931684334511 Recall: 0.6972941176470588 F1 score: 0.622702931501586 Testing Time: 0.001676012487972484 (+/-) 0.0005692564746389187 Training Time: 0.9064657968633315 (+/-) 0.026105526707701558 === Average network evolution === Total hidden node: 6.0588235294117645 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 69.91625 Std Accuracy: 3.004556695670761 Hidden Node mean 5.9875 Hidden Node std: 0.7157819151110204 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (18 of 18) |########################| Elapsed Time: 0:00:07 ETA: 00:00:00C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.6201829209047205 (+/-) 0.02630606436972817 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0016882700078627642 (+/-) 0.0005648622236766069 Training Time: 0.4569523334503174 (+/-) 0.006990880303626607 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:07 ETA: 00:00:00
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.090304173933049 Testing Loss: 0.5974845886230469 (+/-) 0.02406507834589242 Precision: 0.6277162034856335 Recall: 0.686 F1 score: 0.5591066246793293 Testing Time: 0.0015757644877714269 (+/-) 0.0004899215159543164 Training Time: 0.4609805696150836 (+/-) 0.007324266184726144 === Average network evolution === Total hidden node: 4.470588235294118 (+/-) 0.4991341984846218 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:07 ETA: 00:00:00
=== Performance result === Accuracy: 67.90588235294116 (+/-) 4.1766114309309055 Testing Loss: 0.603587971014135 (+/-) 0.025530270864039875 Precision: 0.580652643603637 Recall: 0.6790588235294117 F1 score: 0.568524649612859 Testing Time: 0.001984540153952206 (+/-) 0.0004764899260866398 Training Time: 0.4569560920490938 (+/-) 0.008606815351798011 === Average network evolution === Total hidden node: 6.9411764705882355 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:07 ETA: 00:00:00
=== Performance result === Accuracy: 68.9235294117647 (+/-) 3.6436131730765973 Testing Loss: 0.5910018892849193 (+/-) 0.02375912043887719 Precision: 0.6628754001376249 Recall: 0.6892352941176471 F1 score: 0.5765025082765465 Testing Time: 0.0017439056845272288 (+/-) 0.00041486780139656654 Training Time: 0.46376682730282054 (+/-) 0.005816481142477356 === Average network evolution === Total hidden node: 3.9411764705882355 (+/-) 0.23529411764705882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 3 No. of parameters : 27 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (18 of 18) |########################| Elapsed Time: 0:00:07 ETA: 00:00:00C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.6077858910841101 (+/-) 0.031235013529145882 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0018096110400031595 (+/-) 0.0005030625868849391 Training Time: 0.4586522859685561 (+/-) 0.005943431722970871 === Average network evolution === Total hidden node: 4.705882352941177 (+/-) 0.4556450995538137 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 68.26500000000001 Std Accuracy: 4.026105438261646 Hidden Node mean 4.8375 Hidden Node std: 1.1666592261667499 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
N/A% (0 of 18) | | Elapsed Time: 0:00:00 ETA: --:--:--
=== Performance result === Accuracy: 62.26470588235294 (+/-) 4.457504072971064 Testing Loss: 0.6775683690519894 (+/-) 0.008139513963002729 Precision: 0.46418534062185385 Recall: 0.6226470588235294 F1 score: 0.5288608394023046 Testing Time: 0.001172949286068187 (+/-) 0.000506151804554063 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 18) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.5995138673221364 (+/-) 0.03570022883740612 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0012898865868063534 (+/-) 0.00045586888240667475 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 18) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.6246076787219328 (+/-) 0.020225929634677627 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0011144385618322035 (+/-) 0.0004697886688329447 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 18) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.6027752862257116 (+/-) 0.027897091147674035 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0009980482213637408 (+/-) 0.0005903526480058131 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 18) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 68.60000000000001 (+/-) 4.105806505282918 Testing Loss: 0.6072112146545859 (+/-) 0.031461544096153905 Precision: 0.470596 Recall: 0.686 F1 score: 0.5582396204033215 Testing Time: 0.0012955104603486903 (+/-) 0.0004516335999640168 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 6.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 67.0525 Std Accuracy: 4.883005606181505 Hidden Node mean 5.2 Hidden Node std: 1.1661903789690604 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [4]:
%run NADINE_classification_rfid.ipynb
Number of input: 3 Number of output: 4 Number of batch: 280 All Data
100% (280 of 280) |######################| Elapsed Time: 0:08:20 ETA: 00:00:00
=== Performance result === Accuracy: 98.50752688172042 (+/-) 5.178662520051795 Testing Loss: 0.09580279842993798 (+/-) 0.17195608986540695 Precision: 0.9850633741371198 Recall: 0.9850752688172043 F1 score: 0.9850357947701097 Testing Time: 0.004380185971550617 (+/-) 0.0030120721587800063 Training Time: 1.7873665944649755 (+/-) 0.05223200013442971 === Average network evolution === Total hidden node: 24.03942652329749 (+/-) 5.708751474597842 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=29, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 29 No. of parameters : 116 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=29, out_features=4, bias=True) ) No. of inputs : 29 No. of output : 4 No. of parameters : 120 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:08:19 ETA: 00:00:00
=== Performance result === Accuracy: 98.11612903225807 (+/-) 7.479002247716169 Testing Loss: 0.10154843126760803 (+/-) 0.19792186457054634 Precision: 0.9812388987408331 Recall: 0.9811612903225806 F1 score: 0.9811457087827173 Testing Time: 0.004506550382115081 (+/-) 0.0027239130705414287 Training Time: 1.784056090966775 (+/-) 0.026218185920194703 === Average network evolution === Total hidden node: 25.096774193548388 (+/-) 6.163945082887561 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=30, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 30 No. of parameters : 120 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=30, out_features=4, bias=True) ) No. of inputs : 30 No. of output : 4 No. of parameters : 124 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:08:19 ETA: 00:00:00
=== Performance result === Accuracy: 98.28064516129032 (+/-) 6.4155488438074775 Testing Loss: 0.10940202236122128 (+/-) 0.19013106980344696 Precision: 0.9827822837362377 Recall: 0.9828064516129033 F1 score: 0.9827886087842048 Testing Time: 0.0042245294030849225 (+/-) 0.003184458347739251 Training Time: 1.7818412464579374 (+/-) 0.024887894636771296 === Average network evolution === Total hidden node: 21.232974910394265 (+/-) 5.522147208097677 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=26, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 26 No. of parameters : 104 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=26, out_features=4, bias=True) ) No. of inputs : 26 No. of output : 4 No. of parameters : 108 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:08:20 ETA: 00:00:00
=== Performance result === Accuracy: 98.45949820788529 (+/-) 5.683697471937239 Testing Loss: 0.08817090074061065 (+/-) 0.16505240263601564 Precision: 0.9845958961883633 Recall: 0.9845949820788531 F1 score: 0.9845619718805048 Testing Time: 0.004862935739606084 (+/-) 0.003218650852716822 Training Time: 1.7874196780625211 (+/-) 0.026007434419859567 === Average network evolution === Total hidden node: 27.53405017921147 (+/-) 5.529142901974678 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=32, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 32 No. of parameters : 128 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=32, out_features=4, bias=True) ) No. of inputs : 32 No. of output : 4 No. of parameters : 132 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:08:20 ETA: 00:00:00
=== Performance result === Accuracy: 98.28637992831541 (+/-) 6.305538398396073 Testing Loss: 0.08944008701337387 (+/-) 0.1681688310345236 Precision: 0.9829407479345981 Recall: 0.9828637992831541 F1 score: 0.9828278053312793 Testing Time: 0.004851790739216685 (+/-) 0.004626798879340142 Training Time: 1.784758467828074 (+/-) 0.023383059830753682 === Average network evolution === Total hidden node: 25.3584229390681 (+/-) 5.500132844955836 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=30, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 30 No. of parameters : 120 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=30, out_features=4, bias=True) ) No. of inputs : 30 No. of output : 4 No. of parameters : 124 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 98.54136690647482 Std Accuracy: 5.122668720717178 Hidden Node mean 24.71726618705036 Hidden Node std: 5.961258957099992 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (280 of 280) |######################| Elapsed Time: 0:04:12 ETA: 00:00:00
=== Performance result === Accuracy: 96.81362007168458 (+/-) 10.31655735758229 Testing Loss: 0.1706754369894877 (+/-) 0.2565843639948921 Precision: 0.9681081032009339 Recall: 0.9681362007168459 F1 score: 0.9680816026197927 Testing Time: 0.00449659662007431 (+/-) 0.00370217445706961 Training Time: 0.8991863932660831 (+/-) 0.014419004085141813 === Average network evolution === Total hidden node: 22.351254480286737 (+/-) 6.060500547402051 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=28, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 28 No. of parameters : 112 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=28, out_features=4, bias=True) ) No. of inputs : 28 No. of output : 4 No. of parameters : 116 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:04:13 ETA: 00:00:00
=== Performance result === Accuracy: 96.31039426523297 (+/-) 12.13877637817133 Testing Loss: 0.18882475311343816 (+/-) 0.27630026207253383 Precision: 0.9633266934790227 Recall: 0.9631039426523298 F1 score: 0.9630218263627522 Testing Time: 0.004124692691269741 (+/-) 0.002931483495817556 Training Time: 0.9016470977482403 (+/-) 0.014624521056018365 === Average network evolution === Total hidden node: 18.508960573476703 (+/-) 6.22738995550654 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=25, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 25 No. of parameters : 100 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=25, out_features=4, bias=True) ) No. of inputs : 25 No. of output : 4 No. of parameters : 104 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:04:14 ETA: 00:00:00
=== Performance result === Accuracy: 96.1301075268817 (+/-) 12.078478015513117 Testing Loss: 0.19973284983506767 (+/-) 0.2788525744437138 Precision: 0.9612547351386822 Recall: 0.9613010752688173 F1 score: 0.9612121607371694 Testing Time: 0.0039303841129426035 (+/-) 0.002860409930768772 Training Time: 0.9060161609376203 (+/-) 0.016646389474722852 === Average network evolution === Total hidden node: 17.182795698924732 (+/-) 5.86395264059698 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=23, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 23 No. of parameters : 92 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=23, out_features=4, bias=True) ) No. of inputs : 23 No. of output : 4 No. of parameters : 96 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:04:14 ETA: 00:00:00
=== Performance result === Accuracy: 96.87777777777777 (+/-) 10.189899810310484 Testing Loss: 0.17213533477570633 (+/-) 0.24931515121299216 Precision: 0.9688457636287028 Recall: 0.9687777777777777 F1 score: 0.9686883428727033 Testing Time: 0.004207309428936264 (+/-) 0.0030012260209693418 Training Time: 0.9043187360182458 (+/-) 0.015718687615844565 === Average network evolution === Total hidden node: 19.204301075268816 (+/-) 5.830138860709934 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=25, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 25 No. of parameters : 100 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=25, out_features=4, bias=True) ) No. of inputs : 25 No. of output : 4 No. of parameters : 104 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:04:14 ETA: 00:00:00
=== Performance result === Accuracy: 96.715770609319 (+/-) 11.155356269976496 Testing Loss: 0.18115491464760783 (+/-) 0.26129834564887783 Precision: 0.9671166435898345 Recall: 0.96715770609319 F1 score: 0.9671122038415024 Testing Time: 0.004085470698640338 (+/-) 0.0029432420746605 Training Time: 0.9043886541892978 (+/-) 0.020007712086432053 === Average network evolution === Total hidden node: 18.39068100358423 (+/-) 5.986369155445258 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=24, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 24 No. of parameters : 96 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=24, out_features=4, bias=True) ) No. of inputs : 24 No. of output : 4 No. of parameters : 100 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 96.82784172661871 Std Accuracy: 10.368372374754307 Hidden Node mean 19.176978417266188 Hidden Node std: 6.198309314002664 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (280 of 280) |######################| Elapsed Time: 0:02:10 ETA: 00:00:00
=== Performance result === Accuracy: 93.6189964157706 (+/-) 15.228214866657115 Testing Loss: 0.33884505900858125 (+/-) 0.3348806671384748 Precision: 0.9361037513949562 Recall: 0.9361899641577061 F1 score: 0.9359978299506898 Testing Time: 0.00348150516496337 (+/-) 0.002829386667201641 Training Time: 0.46274802419874406 (+/-) 0.009027895520982626 === Average network evolution === Total hidden node: 13.480286738351255 (+/-) 5.120638393031646 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=20, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 20 No. of parameters : 80 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=20, out_features=4, bias=True) ) No. of inputs : 20 No. of output : 4 No. of parameters : 84 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:02:12 ETA: 00:00:00
=== Performance result === Accuracy: 95.88315412186381 (+/-) 11.154639779832582 Testing Loss: 0.28391984310735513 (+/-) 0.2799758101066172 Precision: 0.9589744902632784 Recall: 0.958831541218638 F1 score: 0.9587437846740416 Testing Time: 0.003650169646013595 (+/-) 0.002533503164186266 Training Time: 0.4680409192184394 (+/-) 0.020443591328717942 === Average network evolution === Total hidden node: 15.189964157706093 (+/-) 4.876588758534216 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=21, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 21 No. of parameters : 84 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=21, out_features=4, bias=True) ) No. of inputs : 21 No. of output : 4 No. of parameters : 88 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:02:10 ETA: 00:00:00
=== Performance result === Accuracy: 94.06236559139785 (+/-) 15.234006813035256 Testing Loss: 0.3155328658105652 (+/-) 0.32502563458095457 Precision: 0.9405964768572362 Recall: 0.9406236559139785 F1 score: 0.9404771155887688 Testing Time: 0.003660654936212793 (+/-) 0.002909406057386138 Training Time: 0.4616563063795849 (+/-) 0.01226758522684207 === Average network evolution === Total hidden node: 15.017921146953405 (+/-) 5.137852498355462 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=21, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 21 No. of parameters : 84 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=21, out_features=4, bias=True) ) No. of inputs : 21 No. of output : 4 No. of parameters : 88 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:02:10 ETA: 00:00:00
=== Performance result === Accuracy: 93.63189964157705 (+/-) 16.38740731823018 Testing Loss: 0.32807887369586575 (+/-) 0.3424237038541921 Precision: 0.9362436201370711 Recall: 0.9363189964157707 F1 score: 0.9362084884709391 Testing Time: 0.003651742012270035 (+/-) 0.0026595894596306508 Training Time: 0.4596922884705246 (+/-) 0.010867386660802052 === Average network evolution === Total hidden node: 13.982078853046595 (+/-) 5.093713320518899 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=20, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 20 No. of parameters : 80 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=20, out_features=4, bias=True) ) No. of inputs : 20 No. of output : 4 No. of parameters : 84 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:02:09 ETA: 00:00:00
=== Performance result === Accuracy: 95.34229390681004 (+/-) 12.643833541993232 Testing Loss: 0.27852294273594375 (+/-) 0.2816820824808159 Precision: 0.9534864187566299 Recall: 0.9534229390681004 F1 score: 0.9533213179337134 Testing Time: 0.003716166301440167 (+/-) 0.002673435005717518 Training Time: 0.45873765364342695 (+/-) 0.009048368249529582 === Average network evolution === Total hidden node: 14.824372759856631 (+/-) 4.773881398085375 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=21, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 21 No. of parameters : 84 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=21, out_features=4, bias=True) ) No. of inputs : 21 No. of output : 4 No. of parameters : 88 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 94.7510071942446 Std Accuracy: 13.727235795406228 Hidden Node mean 14.530935251798562 Hidden Node std: 5.026274117017972 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
100% (280 of 280) |######################| Elapsed Time: 0:00:01 ETA: 00:00:00
=== Performance result === Accuracy: 42.89605734767025 (+/-) 1.4060362230435346 Testing Loss: 1.3622969594053043 (+/-) 0.003979283039759391
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.2261800316595075 Recall: 0.4289605734767025 F1 score: 0.29115950660400347 Testing Time: 0.0024084703042088445 (+/-) 0.0030716220587299367 Training Time: 3.574569592766437e-06 (+/-) 5.959998557920508e-05 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 5 No. of parameters : 20 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=4, bias=True) ) No. of inputs : 5 No. of output : 4 No. of parameters : 24 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:00:01 ETA: 00:00:00
=== Performance result === Accuracy: 50.02329749103943 (+/-) 0.15425350716917904 Testing Loss: 1.2934244260138508 (+/-) 0.0028348320600007737
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.26727256695781665 Recall: 0.5002329749103943 F1 score: 0.343321343876092 Testing Time: 0.002903966493504022 (+/-) 0.002869679805746046 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 10.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 10 No. of parameters : 40 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=4, bias=True) ) No. of inputs : 10 No. of output : 4 No. of parameters : 44 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:00:01 ETA: 00:00:00
=== Performance result === Accuracy: 33.95197132616487 (+/-) 1.2207237053359818 Testing Loss: 1.3639209261931826 (+/-) 0.0029279462833536906
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.25027469373097194 Recall: 0.33951971326164876 F1 score: 0.24837005205321017 Testing Time: 0.002465090016737634 (+/-) 0.0032990957450668426 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 6.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 6 No. of parameters : 24 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=4, bias=True) ) No. of inputs : 6 No. of output : 4 No. of parameters : 28 Dynamic laerning rate for each hidden layer: [0.02]
98% (277 of 280) |##################### | Elapsed Time: 0:00:01 ETA: 0:00:00
=== Performance result === Accuracy: 52.696057347670255 (+/-) 2.143027088175485 Testing Loss: 1.2984814891678458 (+/-) 0.005127954142873944
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.43022215990094465 Recall: 0.5269605734767026 F1 score: 0.4518516442531176 Testing Time: 0.00260938965718806 (+/-) 0.0025831530530461794 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 7 No. of parameters : 28 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=4, bias=True) ) No. of inputs : 7 No. of output : 4 No. of parameters : 32 Dynamic laerning rate for each hidden layer: [0.02]
100% (280 of 280) |######################| Elapsed Time: 0:00:01 ETA: 00:00:00
=== Performance result === Accuracy: 41.3584229390681 (+/-) 2.985231093456183 Testing Loss: 1.3085512362927947 (+/-) 0.008579623438723972
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.4111094070040762 Recall: 0.413584229390681 F1 score: 0.3645665553775302 Testing Time: 0.0028447011038393957 (+/-) 0.0036859425281646698 Training Time: 3.5762786865234375e-06 (+/-) 5.9628481866835596e-05 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 8 No. of parameters : 32 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=4, bias=True) ) No. of inputs : 8 No. of output : 4 No. of parameters : 36 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 44.18553956834533 Std Accuracy: 6.8966574817875514 Hidden Node mean 7.2 Hidden Node std: 1.7204650534085255 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [5]:
%run NADINE_classification_rmnist.ipynb
Number of input: 784 Number of output: 10 Number of batch: 69 All Data
100% (69 of 69) |########################| Elapsed Time: 0:06:51 ETA: 00:00:00
=== Performance result === Accuracy: 89.7779411764706 (+/-) 4.312732366740142 Testing Loss: 0.3656957885798286 (+/-) 0.1914550936991656 Precision: 0.8974027713792336 Recall: 0.8977794117647059 F1 score: 0.8974843421181354 Testing Time: 0.0048507276703329645 (+/-) 0.005889459167104931 Training Time: 6.037652418893926 (+/-) 0.07305288371021151 === Average network evolution === Total hidden node: 19.720588235294116 (+/-) 3.8571766123273536 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=26, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 26 No. of parameters : 20410 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=26, out_features=10, bias=True) ) No. of inputs : 26 No. of output : 10 No. of parameters : 270 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:50 ETA: 00:00:00
=== Performance result === Accuracy: 89.8323529411765 (+/-) 3.920813285099023 Testing Loss: 0.3650701510555604 (+/-) 0.17736439110381158 Precision: 0.8978836143030622 Recall: 0.8983235294117647 F1 score: 0.8979715996898862 Testing Time: 0.0039026491782244515 (+/-) 0.0009169915944134823 Training Time: 6.029626842807321 (+/-) 0.08768680202650103 === Average network evolution === Total hidden node: 21.073529411764707 (+/-) 3.915715607615444 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=27, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 27 No. of parameters : 21195 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=27, out_features=10, bias=True) ) No. of inputs : 27 No. of output : 10 No. of parameters : 280 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:49 ETA: 00:00:00
=== Performance result === Accuracy: 89.39999999999999 (+/-) 4.337422655692853 Testing Loss: 0.37944120311123486 (+/-) 0.19134794980701408 Precision: 0.8935722687340589 Recall: 0.894 F1 score: 0.8935782298699014 Testing Time: 0.004811756751116584 (+/-) 0.006440372476329324 Training Time: 6.020563612965977 (+/-) 0.08327891675695558 === Average network evolution === Total hidden node: 20.397058823529413 (+/-) 3.8848866509407243 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=26, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 26 No. of parameters : 20410 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=26, out_features=10, bias=True) ) No. of inputs : 26 No. of output : 10 No. of parameters : 270 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:46 ETA: 00:00:00
=== Performance result === Accuracy: 89.04558823529413 (+/-) 4.743243920490844 Testing Loss: 0.3986700133365743 (+/-) 0.19186454445809153 Precision: 0.8902575075005407 Recall: 0.8904558823529412 F1 score: 0.8902066074912375 Testing Time: 0.0037770341424381033 (+/-) 0.0007749966947811903 Training Time: 5.977357026408701 (+/-) 0.06471926264580212 === Average network evolution === Total hidden node: 18.191176470588236 (+/-) 3.5614659611440267 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=24, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 24 No. of parameters : 18840 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=24, out_features=10, bias=True) ) No. of inputs : 24 No. of output : 10 No. of parameters : 250 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:48 ETA: 00:00:00
=== Performance result === Accuracy: 89.65735294117647 (+/-) 4.327453951108197 Testing Loss: 0.36609419884488864 (+/-) 0.18164167439906312 Precision: 0.8961433158234535 Recall: 0.8965735294117647 F1 score: 0.8962467389087099 Testing Time: 0.004860043525695801 (+/-) 0.006537807966261404 Training Time: 6.0055275980164025 (+/-) 0.08049486239923274 === Average network evolution === Total hidden node: 22.352941176470587 (+/-) 4.47464985775211 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=29, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 29 No. of parameters : 22765 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=29, out_features=10, bias=True) ) No. of inputs : 29 No. of output : 10 No. of parameters : 300 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 89.8707462686567 Std Accuracy: 3.41583376917049 Hidden Node mean 20.459701492537313 Hidden Node std: 4.110854205509239 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (69 of 69) |########################| Elapsed Time: 0:03:25 ETA: 00:00:00
=== Performance result === Accuracy: 86.54411764705883 (+/-) 6.280323669837716 Testing Loss: 0.48717381476479416 (+/-) 0.25995288275951006 Precision: 0.864775883942096 Recall: 0.8654411764705883 F1 score: 0.8645916353388433 Testing Time: 0.003453356378218707 (+/-) 0.0007120042827385654 Training Time: 3.014961277737337 (+/-) 0.16742799389761504 === Average network evolution === Total hidden node: 14.617647058823529 (+/-) 1.1249759705238398 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=16, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 16 No. of parameters : 12560 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=16, out_features=10, bias=True) ) No. of inputs : 16 No. of output : 10 No. of parameters : 170 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:21 ETA: 00:00:00
=== Performance result === Accuracy: 86.55882352941177 (+/-) 7.188987396872215 Testing Loss: 0.4918177685536006 (+/-) 0.27929160682148796 Precision: 0.8651093602705155 Recall: 0.8655882352941177 F1 score: 0.8650604939743877 Testing Time: 0.004575571593116312 (+/-) 0.0062197466996679615 Training Time: 2.9522136239444507 (+/-) 0.15353610140787435 === Average network evolution === Total hidden node: 17.16176470588235 (+/-) 2.7525939323484288 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=21, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 21 No. of parameters : 16485 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=21, out_features=10, bias=True) ) No. of inputs : 21 No. of output : 10 No. of parameters : 220 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:20 ETA: 00:00:00
=== Performance result === Accuracy: 87.24411764705883 (+/-) 7.165406731876285 Testing Loss: 0.46989954898462577 (+/-) 0.27077468613807293 Precision: 0.8720543288004834 Recall: 0.8724411764705883 F1 score: 0.8716809854999056 Testing Time: 0.0036396033623639274 (+/-) 0.0009791138390918888 Training Time: 2.946575206868789 (+/-) 0.15999828949691033 === Average network evolution === Total hidden node: 16.86764705882353 (+/-) 1.7481453345550657 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=19, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 19 No. of parameters : 14915 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=19, out_features=10, bias=True) ) No. of inputs : 19 No. of output : 10 No. of parameters : 200 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:19 ETA: 00:00:00
=== Performance result === Accuracy: 88.18088235294118 (+/-) 5.507541817227729 Testing Loss: 0.4403886842157911 (+/-) 0.2376126597821049 Precision: 0.8813356969097015 Recall: 0.8818088235294118 F1 score: 0.8812343710882201 Testing Time: 0.004768171731163473 (+/-) 0.007270091557831491 Training Time: 2.923224883921006 (+/-) 0.1555575878756655 === Average network evolution === Total hidden node: 19.455882352941178 (+/-) 2.1586612376888477 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=23, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 23 No. of parameters : 18055 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=23, out_features=10, bias=True) ) No. of inputs : 23 No. of output : 10 No. of parameters : 240 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:21 ETA: 00:00:00
=== Performance result === Accuracy: 87.42352941176469 (+/-) 6.716860209403227 Testing Loss: 0.4575348578612594 (+/-) 0.2581767162026987 Precision: 0.873966356457187 Recall: 0.8742352941176471 F1 score: 0.8737147743115773 Testing Time: 0.0038860264946432676 (+/-) 0.0007411101108326854 Training Time: 2.9637258648872375 (+/-) 0.1902575565004077 === Average network evolution === Total hidden node: 17.926470588235293 (+/-) 1.4981967246011236 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=20, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 20 No. of parameters : 15700 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=20, out_features=10, bias=True) ) No. of inputs : 20 No. of output : 10 No. of parameters : 210 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 87.68985074626865 Std Accuracy: 5.214493433572249 Hidden Node mean 17.256716417910447 Hidden Node std: 2.4749023125963237 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 83.35 (+/-) 9.54306801080363 Testing Loss: 0.6274096062954735 (+/-) 0.33494384504457314 Precision: 0.8333896816792239 Recall: 0.8335 F1 score: 0.8322249567950832 Testing Time: 0.004421023761524874 (+/-) 0.007179156693084881 Training Time: 1.4470381631570703 (+/-) 0.18496766943128867 === Average network evolution === Total hidden node: 14.647058823529411 (+/-) 1.0539101686569952 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=16, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 16 No. of parameters : 12560 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=16, out_features=10, bias=True) ) No. of inputs : 16 No. of output : 10 No. of parameters : 170 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 83.15294117647058 (+/-) 11.028885477639738 Testing Loss: 0.6457117298508391 (+/-) 0.3887588682191189 Precision: 0.831816971391776 Recall: 0.8315294117647059 F1 score: 0.8300960707539232 Testing Time: 0.003357806626488181 (+/-) 0.000703967535306976 Training Time: 1.44128116088755 (+/-) 0.18491195258754953 === Average network evolution === Total hidden node: 14.382352941176471 (+/-) 1.9928505431158734 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:39 ETA: 00:00:00
=== Performance result === Accuracy: 83.82205882352943 (+/-) 8.922754217172475 Testing Loss: 0.6299682180671131 (+/-) 0.3446054502776884 Precision: 0.8379168691773826 Recall: 0.8382205882352941 F1 score: 0.8365749464367875 Testing Time: 0.004208729547612807 (+/-) 0.005610241499072107 Training Time: 1.4526439028627731 (+/-) 0.18853573138181765 === Average network evolution === Total hidden node: 14.588235294117647 (+/-) 0.9886903714100947 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=16, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 16 No. of parameters : 12560 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=16, out_features=10, bias=True) ) No. of inputs : 16 No. of output : 10 No. of parameters : 170 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:39 ETA: 00:00:00
=== Performance result === Accuracy: 84.36911764705881 (+/-) 9.046826830001862 Testing Loss: 0.6119038605076426 (+/-) 0.34689859804912615 Precision: 0.8426964650930981 Recall: 0.8436911764705882 F1 score: 0.8424288780709651 Testing Time: 0.003506635918336756 (+/-) 0.0006323586922781164 Training Time: 1.4529242901241077 (+/-) 0.18544866815226194 === Average network evolution === Total hidden node: 15.102941176470589 (+/-) 1.3841508958282494 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 84.67500000000001 (+/-) 8.374530276555287 Testing Loss: 0.5915400480084559 (+/-) 0.33630332331912927 Precision: 0.8461173734452627 Recall: 0.84675 F1 score: 0.8457453334313139 Testing Time: 0.004223998855142032 (+/-) 0.0060989665834394816 Training Time: 1.4461341465220732 (+/-) 0.18632423017219257 === Average network evolution === Total hidden node: 16.397058823529413 (+/-) 1.086544791886228 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 84.50358208955224 Std Accuracy: 7.9630894217498085 Hidden Node mean 15.056716417910447 Hidden Node std: 1.5179742865103385 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
94% (65 of 69) |###################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 60.69705882352942 (+/-) 4.467923935191537 Testing Loss: 1.5691660221885233 (+/-) 0.06338841915685982 Precision: 0.6598437727719552 Recall: 0.6069705882352942 F1 score: 0.6021829635981736 Testing Time: 0.0024278900202582866 (+/-) 0.0007805335729344191 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 12.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 12 No. of parameters : 9420 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=10, bias=True) ) No. of inputs : 12 No. of output : 10 No. of parameters : 130 Dynamic laerning rate for each hidden layer: [0.02]
95% (66 of 69) |###################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 65.37647058823529 (+/-) 4.270181069554547 Testing Loss: 1.3851431993877186 (+/-) 0.07066970467002194 Precision: 0.7040914077443862 Recall: 0.6537647058823529 F1 score: 0.6409908343494867 Testing Time: 0.0037265700452467974 (+/-) 0.006288660202837343 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 15.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=15, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 15 No. of parameters : 11775 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=15, out_features=10, bias=True) ) No. of inputs : 15 No. of output : 10 No. of parameters : 160 Dynamic laerning rate for each hidden layer: [0.02]
85% (59 of 69) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 65.28235294117648 (+/-) 5.276137769030057 Testing Loss: 1.4826031730455511 (+/-) 0.06977493707502204 Precision: 0.7142942373036114 Recall: 0.6528235294117647 F1 score: 0.6479459439205114 Testing Time: 0.0028160950716804058 (+/-) 0.0007043645295938386 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 14.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 14 No. of parameters : 10990 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=10, bias=True) ) No. of inputs : 14 No. of output : 10 No. of parameters : 150 Dynamic laerning rate for each hidden layer: [0.02]
95% (66 of 69) |###################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 64.83823529411765 (+/-) 4.152994292278223 Testing Loss: 1.4382804264040554 (+/-) 0.06824306673319774 Precision: 0.7150713991808211 Recall: 0.6483823529411765 F1 score: 0.632373266129527 Testing Time: 0.0037043830927680522 (+/-) 0.006854867380029874 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 14.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 14 No. of parameters : 10990 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=10, bias=True) ) No. of inputs : 14 No. of output : 10 No. of parameters : 150 Dynamic laerning rate for each hidden layer: [0.02]
89% (62 of 69) |##################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 69.32794117647059 (+/-) 4.0978418131928365 Testing Loss: 1.4293343915658838 (+/-) 0.06539882975575784 Precision: 0.7169478817581189 Recall: 0.6932794117647059 F1 score: 0.6840222651883603 Testing Time: 0.0027276452849893007 (+/-) 0.0006963476442049866 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 14.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 14 No. of parameters : 10990 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=10, bias=True) ) No. of inputs : 14 No. of output : 10 No. of parameters : 150 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 65.03492537313433 Std Accuracy: 5.240622434586708 Hidden Node mean 13.8 Hidden Node std: 0.9797958971132712 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [6]:
%run NADINE_classification_pmnist.ipynb
Number of input: 784 Number of output: 10 Number of batch: 69 All Data
100% (69 of 69) |########################| Elapsed Time: 0:06:51 ETA: 00:00:00
=== Performance result === Accuracy: 83.43382352941177 (+/-) 13.971195265449865 Testing Loss: 0.546282297328991 (+/-) 0.4328766528412209 Precision: 0.8348656485393544 Recall: 0.8343382352941177 F1 score: 0.8342266216610447 Testing Time: 0.004340568009544821 (+/-) 0.005753344443770464 Training Time: 6.046676106312695 (+/-) 0.12881687404126776 === Average network evolution === Total hidden node: 18.058823529411764 (+/-) 2.5718061457293144 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=22, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 22 No. of parameters : 17270 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=22, out_features=10, bias=True) ) No. of inputs : 22 No. of output : 10 No. of parameters : 230 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:49 ETA: 00:00:00
=== Performance result === Accuracy: 83.1970588235294 (+/-) 14.590024498913385 Testing Loss: 0.5515862648539683 (+/-) 0.44305466265403626 Precision: 0.8319739619568614 Recall: 0.8319705882352941 F1 score: 0.8315980159837275 Testing Time: 0.003988812951480641 (+/-) 0.0008364052576026348 Training Time: 6.018907413763158 (+/-) 0.07880498855890589 === Average network evolution === Total hidden node: 19.602941176470587 (+/-) 3.010470482491326 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=24, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 24 No. of parameters : 18840 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=24, out_features=10, bias=True) ) No. of inputs : 24 No. of output : 10 No. of parameters : 250 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:47 ETA: 00:00:00
=== Performance result === Accuracy: 83.03529411764707 (+/-) 14.379496889524603 Testing Loss: 0.5702499383512665 (+/-) 0.44963685785500546 Precision: 0.8314390468429103 Recall: 0.8303529411764706 F1 score: 0.8304347357291668 Testing Time: 0.004570813740001005 (+/-) 0.006556913438598294 Training Time: 5.9818368448930626 (+/-) 0.10087611624547048 === Average network evolution === Total hidden node: 15.617647058823529 (+/-) 1.645219561287882 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:06:46 ETA: 00:00:00
=== Performance result === Accuracy: 83.12205882352941 (+/-) 14.284646689903232 Testing Loss: 0.5597732691642117 (+/-) 0.42725717543399905 Precision: 0.832537545123628 Recall: 0.8312205882352941 F1 score: 0.8314141384599958 Testing Time: 0.0037112165899837717 (+/-) 0.0007631935862286341 Training Time: 5.9789870065801285 (+/-) 0.05887288300015947 === Average network evolution === Total hidden node: 17.647058823529413 (+/-) 3.3727485121116807 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=22, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 22 No. of parameters : 17270 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=22, out_features=10, bias=True) ) No. of inputs : 22 No. of output : 10 No. of parameters : 230 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:07:02 ETA: 00:00:00
=== Performance result === Accuracy: 77.13970588235296 (+/-) 12.908036025327084 Testing Loss: 0.7634805611827794 (+/-) 0.3704979447579078 Precision: 0.7697418307483593 Recall: 0.7713970588235294 F1 score: 0.7697300437761613 Testing Time: 0.005279881112715777 (+/-) 0.007035512056852779 Training Time: 6.211735024171717 (+/-) 1.157587575913904 === Average network evolution === Total hidden node: 23.794117647058822 (+/-) 7.899290406570431 Number of layer: 1.4558823529411764 (+/-) 0.4980498300551794 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=18, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 18 No. of parameters : 14130 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=15, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 15 No. of parameters : 285 3 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=15, out_features=10, bias=True) ) No. of inputs : 15 No. of output : 10 No. of parameters : 160 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001]
Mean Accuracy: 82.87134328358208 Std Accuracy: 12.353004939342092 Hidden Node mean 19.035820895522388 Hidden Node std: 5.073069119998247 Hidden Layer mean: 1.0925373134328358 Hidden Layer std: 0.28978295163012774 50% Data
100% (69 of 69) |########################| Elapsed Time: 0:03:30 ETA: 00:00:00
=== Performance result === Accuracy: 79.2014705882353 (+/-) 16.28436179326168 Testing Loss: 0.6970983947462895 (+/-) 0.4724680205227353 Precision: 0.7925648554943281 Recall: 0.7920147058823529 F1 score: 0.7915171975795074 Testing Time: 0.004289150238037109 (+/-) 0.007062555281055703 Training Time: 3.090509425191318 (+/-) 0.2580193687585982 === Average network evolution === Total hidden node: 13.426470588235293 (+/-) 0.9125944990647178 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 14 No. of parameters : 10990 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=10, bias=True) ) No. of inputs : 14 No. of output : 10 No. of parameters : 150 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:22 ETA: 00:00:00
=== Performance result === Accuracy: 79.81911764705883 (+/-) 15.527577529265008 Testing Loss: 0.6885084801298731 (+/-) 0.48979780889463453 Precision: 0.7976199672398673 Recall: 0.7981911764705882 F1 score: 0.7973360027023904 Testing Time: 0.0034692427691291362 (+/-) 0.0007128913802272943 Training Time: 2.9678158584763024 (+/-) 0.1629995484955209 === Average network evolution === Total hidden node: 14.573529411764707 (+/-) 1.1156134750017748 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=16, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 16 No. of parameters : 12560 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=16, out_features=10, bias=True) ) No. of inputs : 16 No. of output : 10 No. of parameters : 170 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:20 ETA: 00:00:00
=== Performance result === Accuracy: 80.30588235294117 (+/-) 16.09669204556292 Testing Loss: 0.6618845344466322 (+/-) 0.46388060327505837 Precision: 0.8027461312808202 Recall: 0.8030588235294117 F1 score: 0.8021167475459899 Testing Time: 0.004239040262558881 (+/-) 0.006098644535893181 Training Time: 2.9429731298895443 (+/-) 0.15160792644656734 === Average network evolution === Total hidden node: 15.264705882352942 (+/-) 2.30483919981942 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=19, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 19 No. of parameters : 14915 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=19, out_features=10, bias=True) ) No. of inputs : 19 No. of output : 10 No. of parameters : 200 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:20 ETA: 00:00:00
=== Performance result === Accuracy: 80.52352941176471 (+/-) 15.801276173769104 Testing Loss: 0.6534879685324781 (+/-) 0.470075548367028 Precision: 0.8052034206139526 Recall: 0.805235294117647 F1 score: 0.8041090727520479 Testing Time: 0.0036011583664838005 (+/-) 0.0008115042590571057 Training Time: 2.9495151183184456 (+/-) 0.15283978600116885 === Average network evolution === Total hidden node: 16.779411764705884 (+/-) 1.7050581173059773 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=18, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 18 No. of parameters : 14130 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=18, out_features=10, bias=True) ) No. of inputs : 18 No. of output : 10 No. of parameters : 190 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:03:23 ETA: 00:00:00
=== Performance result === Accuracy: 80.13823529411764 (+/-) 16.693350467386395 Testing Loss: 0.6653229368521887 (+/-) 0.464202628984187 Precision: 0.800497870797109 Recall: 0.8013823529411764 F1 score: 0.8001602451166178 Testing Time: 0.004378034788019517 (+/-) 0.006396727333958313 Training Time: 2.9913469973732445 (+/-) 0.18330792891265285 === Average network evolution === Total hidden node: 16.13235294117647 (+/-) 2.035742470167651 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=19, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 19 No. of parameters : 14915 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=19, out_features=10, bias=True) ) No. of inputs : 19 No. of output : 10 No. of parameters : 200 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 80.91164179104479 Std Accuracy: 14.348879471256337 Hidden Node mean 15.274626865671642 Hidden Node std: 2.052108001258821 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 74.87794117647059 (+/-) 18.44706931600467 Testing Loss: 0.8928117296274971 (+/-) 0.5125161033022954 Precision: 0.7490269541578192 Recall: 0.7487794117647059 F1 score: 0.7467287349254725 Testing Time: 0.003241721321554745 (+/-) 0.0006923762286360324 Training Time: 1.4421351306578691 (+/-) 0.18401023711539968 === Average network evolution === Total hidden node: 12.117647058823529 (+/-) 0.6308120761625652 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 13 No. of parameters : 10205 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=10, bias=True) ) No. of inputs : 13 No. of output : 10 No. of parameters : 140 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 76.87352941176471 (+/-) 17.054261306773835 Testing Loss: 0.8257875142290312 (+/-) 0.5118861190593397 Precision: 0.7691152618871726 Recall: 0.7687352941176471 F1 score: 0.7672772472609044 Testing Time: 0.004467497853671803 (+/-) 0.006342806043401021 Training Time: 1.4415201860315658 (+/-) 0.18527560243828003 === Average network evolution === Total hidden node: 16.25 (+/-) 1.6122235285103468 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=18, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 18 No. of parameters : 14130 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=18, out_features=10, bias=True) ) No. of inputs : 18 No. of output : 10 No. of parameters : 190 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 76.07941176470588 (+/-) 17.87125163350617 Testing Loss: 0.8214778891381096 (+/-) 0.5014818384573742 Precision: 0.762743532432833 Recall: 0.7607941176470588 F1 score: 0.75942443611301 Testing Time: 0.0036011303172392003 (+/-) 0.0006526407862191999 Training Time: 1.4488479915787191 (+/-) 0.18591746361674685 === Average network evolution === Total hidden node: 15.25 (+/-) 0.7928912098063865 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:39 ETA: 00:00:00
=== Performance result === Accuracy: 75.99705882352943 (+/-) 17.30502282225865 Testing Loss: 0.8297660931506577 (+/-) 0.4981708867358865 Precision: 0.759132784843177 Recall: 0.7599705882352941 F1 score: 0.7574713918203102 Testing Time: 0.004442327162798713 (+/-) 0.0071704439128577515 Training Time: 1.4539881453794592 (+/-) 0.18805963960128272 === Average network evolution === Total hidden node: 15.455882352941176 (+/-) 1.2418418902085793 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
100% (69 of 69) |########################| Elapsed Time: 0:01:38 ETA: 00:00:00
=== Performance result === Accuracy: 75.72058823529412 (+/-) 17.78427559608512 Testing Loss: 0.8550452355514554 (+/-) 0.5045557174402614 Precision: 0.7580121618624277 Recall: 0.7572058823529412 F1 score: 0.7548307913460777 Testing Time: 0.003461006809683407 (+/-) 0.0006512643452812632 Training Time: 1.440460040288813 (+/-) 0.1854585688186084 === Average network evolution === Total hidden node: 14.5 (+/-) 1.8510728208893603 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=17, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 17 No. of parameters : 13345 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=17, out_features=10, bias=True) ) No. of inputs : 17 No. of output : 10 No. of parameters : 180 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 76.79671641791045 Std Accuracy: 16.272928383462904 Hidden Node mean 14.73134328358209 Hidden Node std: 1.933074957833799 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
85% (59 of 69) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 23.248529411764707 (+/-) 26.190189522970908 Testing Loss: 2.0830139149637783 (+/-) 0.4251529294151486 Precision: 0.4797908331125918 Recall: 0.23248529411764707 F1 score: 0.247033387642271 Testing Time: 0.003578028258155374 (+/-) 0.007281525541163757 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 14.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 14 No. of parameters : 10990 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=10, bias=True) ) No. of inputs : 14 No. of output : 10 No. of parameters : 150 Dynamic laerning rate for each hidden layer: [0.02]
86% (60 of 69) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 23.46764705882353 (+/-) 25.61873419048044 Testing Loss: 2.090025575721965 (+/-) 0.38214799214450546 Precision: 0.43552230009184817 Recall: 0.2346764705882353 F1 score: 0.24772504730411182 Testing Time: 0.002779673127567067 (+/-) 0.0007856418923005174 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 13.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 13 No. of parameters : 10205 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=10, bias=True) ) No. of inputs : 13 No. of output : 10 No. of parameters : 140 Dynamic laerning rate for each hidden layer: [0.02]
86% (60 of 69) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 24.14117647058823 (+/-) 20.631281469809508 Testing Loss: 2.087069686721353 (+/-) 0.355490693508642 Precision: 0.4661501442132742 Recall: 0.24141176470588235 F1 score: 0.2430880855219304 Testing Time: 0.003593287047217874 (+/-) 0.006063696645095653 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 12.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 12 No. of parameters : 9420 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=10, bias=True) ) No. of inputs : 12 No. of output : 10 No. of parameters : 130 Dynamic laerning rate for each hidden layer: [0.02]
92% (64 of 69) |###################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 24.51617647058824 (+/-) 21.474592444904292 Testing Loss: 2.1386852352058185 (+/-) 0.346134032465388 Precision: 0.49766362132335695 Recall: 0.24516176470588236 F1 score: 0.23431927755941334 Testing Time: 0.0025979210348690256 (+/-) 0.0007097978862626402 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 12.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 12 No. of parameters : 9420 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=10, bias=True) ) No. of inputs : 12 No. of output : 10 No. of parameters : 130 Dynamic laerning rate for each hidden layer: [0.02]
86% (60 of 69) |#################### | Elapsed Time: 0:00:00 ETA: 0:00:00
=== Performance result === Accuracy: 19.075 (+/-) 22.934749995431414 Testing Loss: 2.1456482042284573 (+/-) 0.3726709008797878 Precision: 0.3988332615779539 Recall: 0.19075 F1 score: 0.18674455599811174 Testing Time: 0.003675239927628461 (+/-) 0.006296355685882603 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 13.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=784, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 784 No. of nodes : 13 No. of parameters : 10205 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=10, bias=True) ) No. of inputs : 13 No. of output : 10 No. of parameters : 140 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 22.88059701492537 Std Accuracy: 23.729876871993902 Hidden Node mean 12.8 Hidden Node std: 0.7483314773547882 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [7]:
%run NADINE_classification_hepmass.ipynb
Number of input: 28 Number of output: 2 Number of batch: 2000 All Data
100% (2000 of 2000) |####################| Elapsed Time: 0:59:52 ETA: 00:00:00
=== Performance result === Accuracy: 83.78509254627313 (+/-) 1.6814581557994008 Testing Loss: 0.33565192985319986 (+/-) 0.026867821059909737 Precision: 0.8391347537967414 Recall: 0.8378509254627313 F1 score: 0.8376962830723401 Testing Time: 0.009975884663217839 (+/-) 0.005409544062119003 Training Time: 1.7738356884865238 (+/-) 0.02726461092370268 === Average network evolution === Total hidden node: 4.983991995997999 (+/-) 0.24652435795576202 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 5 No. of parameters : 145 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:59:56 ETA: 00:00:00
=== Performance result === Accuracy: 83.93961980990494 (+/-) 1.6182275444821175 Testing Loss: 0.33559928182186394 (+/-) 0.027063566782921152 Precision: 0.8412817396643667 Recall: 0.8393961980990495 F1 score: 0.8391728227439491 Testing Time: 0.009880155608199607 (+/-) 0.005521819304928295 Training Time: 1.775860794786813 (+/-) 0.025612896446960798 === Average network evolution === Total hidden node: 3.944472236118059 (+/-) 0.2746924408735865 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 4 No. of parameters : 116 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 1:00:14 ETA: 00:00:00
=== Performance result === Accuracy: 84.11960980490245 (+/-) 1.6797697642993932 Testing Loss: 0.3324760759604341 (+/-) 0.026801875157202868 Precision: 0.8426674639399709 Recall: 0.8411960980490245 F1 score: 0.8410243671278081 Testing Time: 0.010260284394249431 (+/-) 0.005705364964115679 Training Time: 1.7843098704847113 (+/-) 0.033449489504207744 === Average network evolution === Total hidden node: 5.901950975487744 (+/-) 0.37478183828856276 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 6 No. of parameters : 174 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 1:00:02 ETA: 00:00:00
=== Performance result === Accuracy: 84.19574787393697 (+/-) 1.631541569317324 Testing Loss: 0.33170405247916335 (+/-) 0.02583321931339422 Precision: 0.8433637654415809 Recall: 0.8419574787393697 F1 score: 0.8417944466392715 Testing Time: 0.01040497918675219 (+/-) 0.005742835462958587 Training Time: 1.77863155155554 (+/-) 0.034280038986008565 === Average network evolution === Total hidden node: 6.877938969484743 (+/-) 0.3580120676701238 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 7 No. of parameters : 203 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:59:37 ETA: 00:00:00
=== Performance result === Accuracy: 84.30820410205101 (+/-) 1.524068278792782 Testing Loss: 0.3297780630527704 (+/-) 0.025375605231254043 Precision: 0.8448926775776725 Recall: 0.8430820410205102 F1 score: 0.8428746483999315 Testing Time: 0.00905679571085897 (+/-) 0.0054888112306791985 Training Time: 1.767761875832898 (+/-) 0.21528079700269565 === Average network evolution === Total hidden node: 6.296648324162081 (+/-) 0.477134782515681 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 6 No. of parameters : 174 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 84.08202202202202 Std Accuracy: 1.5405656292328116 Hidden Node mean 5.601801801801802 Hidden Node std: 1.0914155692042598 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (2000 of 2000) |####################| Elapsed Time: 0:27:02 ETA: 00:00:00
=== Performance result === Accuracy: 84.07928964482241 (+/-) 1.7856527023569115 Testing Loss: 0.3338425622515943 (+/-) 0.028417253480859106 Precision: 0.8422719072357879 Recall: 0.8407928964482241 F1 score: 0.8406196364678606 Testing Time: 0.007952979709459221 (+/-) 0.004475836587480246 Training Time: 0.7922585656966132 (+/-) 0.04383500680962723 === Average network evolution === Total hidden node: 6.314657328664333 (+/-) 0.5016633364983542 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 6 No. of parameters : 174 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:42 ETA: 00:00:00
=== Performance result === Accuracy: 83.85717858929465 (+/-) 1.655704660434071 Testing Loss: 0.33838523768555706 (+/-) 0.02798390451082156 Precision: 0.8401650434484994 Recall: 0.8385717858929465 F1 score: 0.8383813919271698 Testing Time: 0.007975316035741564 (+/-) 0.004470240770647941 Training Time: 0.78202047975377 (+/-) 0.021304375886472562 === Average network evolution === Total hidden node: 6.0605302651325665 (+/-) 0.44602397654532705 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 6 No. of parameters : 174 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:43 ETA: 00:00:00
=== Performance result === Accuracy: 83.94857428714357 (+/-) 1.65074504329353 Testing Loss: 0.33611741885654683 (+/-) 0.028034849815113155 Precision: 0.8411678174663348 Recall: 0.8394857428714357 F1 score: 0.8392864810134615 Testing Time: 0.007847910227925853 (+/-) 0.004792904192226627 Training Time: 0.7828082897592747 (+/-) 0.02103366472386999 === Average network evolution === Total hidden node: 4.540770385192596 (+/-) 0.507289146288754 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 4 No. of parameters : 116 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:36 ETA: 00:00:00
=== Performance result === Accuracy: 83.41745872936468 (+/-) 2.059284018902648 Testing Loss: 0.34196616441622685 (+/-) 0.03507304970696744 Precision: 0.8361367443950527 Recall: 0.8341745872936468 F1 score: 0.8339309098923363 Testing Time: 0.00787491426281836 (+/-) 0.004789848638469876 Training Time: 0.7789903481403788 (+/-) 0.015685693785282055 === Average network evolution === Total hidden node: 4.888444222111056 (+/-) 0.4540535354690224 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 5 No. of parameters : 145 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:45 ETA: 00:00:00
=== Performance result === Accuracy: 83.84632316158078 (+/-) 1.6779520252731805 Testing Loss: 0.33673544112535164 (+/-) 0.027558521239185865 Precision: 0.8402687536777249 Recall: 0.8384632316158079 F1 score: 0.8382474381771258 Testing Time: 0.00798410949497118 (+/-) 0.004686535478036499 Training Time: 0.7836397167680978 (+/-) 0.02311818163678738 === Average network evolution === Total hidden node: 6.065532766383192 (+/-) 0.726272221327263 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 5 No. of parameters : 145 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 83.84343343343342 Std Accuracy: 1.6752297806991925 Hidden Node mean 5.574474474474474 Hidden Node std: 0.8948798293671095 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (2000 of 2000) |####################| Elapsed Time: 0:14:12 ETA: 00:00:00
=== Performance result === Accuracy: 83.47398699349675 (+/-) 2.499399359928525 Testing Loss: 0.3441110202942925 (+/-) 0.03773954841893645 Precision: 0.836080298928544 Recall: 0.8347398699349675 F1 score: 0.8345738307554551 Testing Time: 0.007979063703871895 (+/-) 0.004515432249529046 Training Time: 0.4070171101681288 (+/-) 0.013130394114696834 === Average network evolution === Total hidden node: 5.179589794897448 (+/-) 0.4531736035708553 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 5 No. of parameters : 145 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:12 ETA: 00:00:00
=== Performance result === Accuracy: 83.53186593296648 (+/-) 2.0733184499088373 Testing Loss: 0.3437170646767905 (+/-) 0.03408279775211389 Precision: 0.8367160011872478 Recall: 0.8353186593296649 F1 score: 0.835146518768847 Testing Time: 0.008240557480240059 (+/-) 0.004648430910990265 Training Time: 0.40652117555054385 (+/-) 0.01189760135514526 === Average network evolution === Total hidden node: 7.039019509754877 (+/-) 0.19364138920532062 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 7 No. of parameters : 203 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:14 ETA: 00:00:00
=== Performance result === Accuracy: 83.51310655327664 (+/-) 2.0845601453668596 Testing Loss: 0.3438651190959793 (+/-) 0.03423983403802886 Precision: 0.8364595259580453 Recall: 0.8351310655327664 F1 score: 0.8349670803803341 Testing Time: 0.008017756331855502 (+/-) 0.004690688222398447 Training Time: 0.40793715303334194 (+/-) 0.01452082886345455 === Average network evolution === Total hidden node: 5.992496248124062 (+/-) 0.13210771394953794 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 6 No. of parameters : 174 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:14 ETA: 00:00:00
=== Performance result === Accuracy: 83.42786393196597 (+/-) 1.804612784822916 Testing Loss: 0.34438036775040354 (+/-) 0.03297467971465163 Precision: 0.8360139789263715 Recall: 0.8342786393196598 F1 score: 0.8340631485023473 Testing Time: 0.0076815375928702264 (+/-) 0.00468026620814667 Training Time: 0.40795423496717687 (+/-) 0.015023772617767375 === Average network evolution === Total hidden node: 3.372186093046523 (+/-) 0.6765912068725857 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 4 No. of parameters : 116 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:19 ETA: 00:00:00
=== Performance result === Accuracy: 83.47393696848425 (+/-) 2.167656666942873 Testing Loss: 0.3458197766420184 (+/-) 0.03564322691972024 Precision: 0.8359443878998468 Recall: 0.8347393696848424 F1 score: 0.8345900042082209 Testing Time: 0.00797703410459197 (+/-) 0.004787995842104333 Training Time: 0.4101819817932801 (+/-) 0.01459903164948159 === Average network evolution === Total hidden node: 4.6968484242121065 (+/-) 0.5747262197343124 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 5 No. of parameters : 145 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 83.49955955955957 Std Accuracy: 2.023120565618677 Hidden Node mean 5.256256256256257 Hidden Node std: 1.3142776434690353 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 infinite delay
100% (2000 of 2000) |####################| Elapsed Time: 0:00:37 ETA: 00:00:00
=== Performance result === Accuracy: 69.91825912956477 (+/-) 1.4685571493598437 Testing Loss: 0.6425071584397164 (+/-) 0.003226252701281 Precision: 0.7258352718462668 Recall: 0.6991825912956479 F1 score: 0.6900518192228281 Testing Time: 0.007195837560923711 (+/-) 0.004749730449307384 Training Time: 9.973267425949303e-07 (+/-) 3.151457822259044e-05 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 4 No. of parameters : 116 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:37 ETA: 00:00:00
=== Performance result === Accuracy: 62.393896948474236 (+/-) 1.5286042937690438 Testing Loss: 0.6590038358777568 (+/-) 0.0038654162989762027 Precision: 0.6468573262767628 Recall: 0.6239389694847424 F1 score: 0.6086421519551175 Testing Time: 0.007115222860301 (+/-) 0.004801325691911151 Training Time: 4.985441023734523e-07 (+/-) 2.2284419480405576e-05 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 3 No. of parameters : 87 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:37 ETA: 00:00:00
=== Performance result === Accuracy: 58.016758379189596 (+/-) 1.555604142856791 Testing Loss: 0.6851458680098984 (+/-) 0.0017240404671653093 Precision: 0.5962084719516195 Recall: 0.5801675837918959 F1 score: 0.561858701726377 Testing Time: 0.007289949687139102 (+/-) 0.005053250819275446 Training Time: 2.0082501425273183e-06 (+/-) 4.485227638118325e-05 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 3 No. of parameters : 87 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
99% (1999 of 2000) |################### | Elapsed Time: 0:00:39 ETA: 0:00:00
=== Performance result === Accuracy: 50.22846423211605 (+/-) 1.6139633302562995 Testing Loss: 0.681609387812822 (+/-) 0.007060664959630253 Precision: 0.7486344508425445 Recall: 0.5022846423211605 F1 score: 0.33831337782054066 Testing Time: 0.007450149797570294 (+/-) 0.0048296605113526715 Training Time: 1.4971828031325232e-06 (+/-) 3.861842050085102e-05 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 3 No. of parameters : 87 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:37 ETA: 00:00:00
=== Performance result === Accuracy: 64.38759379689844 (+/-) 1.5414682044058314 Testing Loss: 0.6761737258330531 (+/-) 0.0015618760469291468 Precision: 0.6448363496820557 Recall: 0.6438759379689845 F1 score: 0.6432791386246188 Testing Time: 0.007348885948864325 (+/-) 0.004776797935159196 Training Time: 9.978038182909815e-07 (+/-) 3.1529656969174065e-05 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=28, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 28 No. of nodes : 3 No. of parameters : 87 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 60.98926926926927 Std Accuracy: 6.778579505078718 Hidden Node mean 3.2 Hidden Node std: 0.4 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [8]:
%run NADINE_classification_susy.ipynb
Number of input: 18 Number of output: 2 Number of batch: 2000 All Data
100% (2000 of 2000) |####################| Elapsed Time: 0:51:42 ETA: 00:00:00
=== Performance result === Accuracy: 77.95457728864433 (+/-) 2.5271389008949745 Testing Loss: 0.46798808101178885 (+/-) 0.03389240952010673 Precision: 0.7816274925294973 Recall: 0.7795457728864432 F1 score: 0.7774906867118454 Testing Time: 0.008336776491998136 (+/-) 0.004408220289888391 Training Time: 1.531872933479832 (+/-) 0.03883922979344989 === Average network evolution === Total hidden node: 10.646823411705853 (+/-) 1.9365442845995733 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 12 No. of parameters : 228 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:51:41 ETA: 00:00:00
=== Performance result === Accuracy: 78.08119059529764 (+/-) 2.5774091452741192 Testing Loss: 0.4665262839417269 (+/-) 0.03355245523942899 Precision: 0.7829252115003457 Recall: 0.7808119059529764 F1 score: 0.7787698326157988 Testing Time: 0.008957786521892538 (+/-) 0.004780694020687537 Training Time: 1.5309998615793492 (+/-) 0.029274536249909736 === Average network evolution === Total hidden node: 18.209104552276138 (+/-) 2.175008616032174 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=20, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 20 No. of parameters : 380 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=20, out_features=2, bias=True) ) No. of inputs : 20 No. of output : 2 No. of parameters : 42 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:51:28 ETA: 00:00:00
=== Performance result === Accuracy: 77.73356678339171 (+/-) 2.7136508221968447 Testing Loss: 0.4739224623148891 (+/-) 0.0357267123254208 Precision: 0.7794798124049296 Recall: 0.7773356678339169 F1 score: 0.7752049513975948 Testing Time: 0.008148812007283854 (+/-) 0.004738067416356157 Training Time: 1.5253643019906635 (+/-) 0.03378511922916777 === Average network evolution === Total hidden node: 8.349174587293646 (+/-) 2.15510099512392 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 10 No. of parameters : 190 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=2, bias=True) ) No. of inputs : 10 No. of output : 2 No. of parameters : 22 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:51:40 ETA: 00:00:00
=== Performance result === Accuracy: 77.77358679339669 (+/-) 2.72511941559637 Testing Loss: 0.4718902396553454 (+/-) 0.036043795504685075 Precision: 0.77929750681394 Recall: 0.777735867933967 F1 score: 0.775883359376995 Testing Time: 0.008288919597700155 (+/-) 0.004707945583703868 Training Time: 1.5312942700006773 (+/-) 0.03239534576375941 === Average network evolution === Total hidden node: 10.02751375687844 (+/-) 2.284152369635112 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 12 No. of parameters : 228 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:51:31 ETA: 00:00:00
=== Performance result === Accuracy: 77.61080540270135 (+/-) 2.8062110954108257 Testing Loss: 0.4758321739930758 (+/-) 0.03581825398119729 Precision: 0.7796128835697405 Recall: 0.7761080540270135 F1 score: 0.7733772661883147 Testing Time: 0.007715155805212787 (+/-) 0.00491106841488459 Training Time: 1.5273227577152224 (+/-) 0.024552974930873876 === Average network evolution === Total hidden node: 4.882941470735368 (+/-) 1.828927729681038 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 6 No. of parameters : 114 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 77.8421121121121 Std Accuracy: 2.6280654618016954 Hidden Node mean 10.425225225225224 Hidden Node std: 4.847541953125485 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 50% Data
100% (2000 of 2000) |####################| Elapsed Time: 0:26:40 ETA: 00:00:00
=== Performance result === Accuracy: 77.38724362181091 (+/-) 2.983360953265096 Testing Loss: 0.47770371718547416 (+/-) 0.03861482503176204 Precision: 0.7762272777490693 Recall: 0.7738724362181091 F1 score: 0.7715685100925419 Testing Time: 0.008478721777995626 (+/-) 0.004863724498605859 Training Time: 0.7805007695555389 (+/-) 0.022941870370947432 === Average network evolution === Total hidden node: 11.347673836918458 (+/-) 2.416320062200439 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 14 No. of parameters : 266 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=2, bias=True) ) No. of inputs : 14 No. of output : 2 No. of parameters : 30 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:27 ETA: 00:00:00
=== Performance result === Accuracy: 77.06058029014507 (+/-) 3.1176749047708032 Testing Loss: 0.4833156440692165 (+/-) 0.03914441432304305 Precision: 0.7725210850586511 Recall: 0.7706058029014508 F1 score: 0.7684303150450307 Testing Time: 0.007840401533545703 (+/-) 0.004743831069515042 Training Time: 0.7747281516057483 (+/-) 0.015713147162417104 === Average network evolution === Total hidden node: 6.172586293146574 (+/-) 2.1288482477300135 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 9 No. of parameters : 171 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:35 ETA: 00:00:00
=== Performance result === Accuracy: 76.64967483741871 (+/-) 4.192609883251568 Testing Loss: 0.48823821927798633 (+/-) 0.04787508510173317 Precision: 0.7693834489704106 Recall: 0.7664967483741871 F1 score: 0.7637658827575643 Testing Time: 0.008113261042027666 (+/-) 0.004872810113813746 Training Time: 0.7783785801401372 (+/-) 0.021566474492765216 === Average network evolution === Total hidden node: 6.6368184092046025 (+/-) 2.188277715972628 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 9 No. of parameters : 171 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:30 ETA: 00:00:00
=== Performance result === Accuracy: 76.93026513256629 (+/-) 3.600507090035036 Testing Loss: 0.4850121295529643 (+/-) 0.0438590268283625 Precision: 0.7728522333714393 Recall: 0.7693026513256628 F1 score: 0.7663638575201127 Testing Time: 0.007722694316823939 (+/-) 0.004940909900269146 Training Time: 0.7763779255197667 (+/-) 0.018430621292625567 === Average network evolution === Total hidden node: 5.182591295647824 (+/-) 1.9670057560582925 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 8 No. of parameters : 152 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:26:32 ETA: 00:00:00
=== Performance result === Accuracy: 77.23876938469236 (+/-) 3.387905198963182 Testing Loss: 0.47993546693011363 (+/-) 0.04092340486534516 Precision: 0.7741033372285039 Recall: 0.7723876938469234 F1 score: 0.7703491938942769 Testing Time: 0.008302591394459742 (+/-) 0.004655867523796111 Training Time: 0.7770098834827341 (+/-) 0.018997835725836965 === Average network evolution === Total hidden node: 10.449224612306153 (+/-) 2.4230628893938175 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 13 No. of parameters : 247 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=2, bias=True) ) No. of inputs : 13 No. of output : 2 No. of parameters : 28 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 77.06343343343343 Std Accuracy: 3.4630370845794443 Hidden Node mean 7.95965965965966 Hidden Node std: 3.323178870010817 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (2000 of 2000) |####################| Elapsed Time: 0:14:07 ETA: 00:00:00
=== Performance result === Accuracy: 74.45577788894447 (+/-) 5.171462790776855 Testing Loss: 0.5183526440344196 (+/-) 0.056589860443886884 Precision: 0.7487126942721679 Recall: 0.7445577788894447 F1 score: 0.7404975105007559 Testing Time: 0.006692481315273115 (+/-) 0.003870272721716512 Training Time: 0.4058427988379642 (+/-) 0.020509090982460655 === Average network evolution === Total hidden node: 2.8099049524762383 (+/-) 0.7184288489307575 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 4 No. of parameters : 76 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:06 ETA: 00:00:00
=== Performance result === Accuracy: 76.06213106553277 (+/-) 3.636414912731932 Testing Loss: 0.4964904447774281 (+/-) 0.04589634448868197 Precision: 0.7632474861248275 Recall: 0.7606213106553277 F1 score: 0.7578476213814805 Testing Time: 0.008248140002084172 (+/-) 0.0045581033909403434 Training Time: 0.40368189449129016 (+/-) 0.011363983131302597 === Average network evolution === Total hidden node: 7.67183591795898 (+/-) 1.6120364486718248 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 10 No. of parameters : 190 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=2, bias=True) ) No. of inputs : 10 No. of output : 2 No. of parameters : 22 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:01 ETA: 00:00:00
=== Performance result === Accuracy: 75.2928464232116 (+/-) 4.008896291582547 Testing Loss: 0.5105746760405321 (+/-) 0.047364764191401024 Precision: 0.7571111846534918 Recall: 0.752928464232116 F1 score: 0.74917750993114 Testing Time: 0.00743556797892049 (+/-) 0.004566097001783393 Training Time: 0.40202717783452274 (+/-) 0.011475572168860675 === Average network evolution === Total hidden node: 3.5602801400700352 (+/-) 0.9298792713521609 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 5 No. of parameters : 95 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:07 ETA: 00:00:00
=== Performance result === Accuracy: 75.7591795897949 (+/-) 4.217196568896687 Testing Loss: 0.5010643777190118 (+/-) 0.048502255333986555 Precision: 0.7593131910056503 Recall: 0.757591795897949 F1 score: 0.7552004279454524 Testing Time: 0.007932956603958107 (+/-) 0.004618080317077822 Training Time: 0.4044680723015698 (+/-) 0.015074947785046549 === Average network evolution === Total hidden node: 6.427213606803401 (+/-) 1.4503901563709665 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 9 No. of parameters : 171 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:14:00 ETA: 00:00:00
=== Performance result === Accuracy: 75.06438219109555 (+/-) 5.067929948508019 Testing Loss: 0.5097242912958717 (+/-) 0.052509488367184984 Precision: 0.7543043343139796 Recall: 0.7506438219109555 F1 score: 0.7470539921002298 Testing Time: 0.006962647492913022 (+/-) 0.004383670312037916 Training Time: 0.40201500608302043 (+/-) 0.011512957053012264 === Average network evolution === Total hidden node: 3.011505752876438 (+/-) 0.6705165770374076 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 4 No. of parameters : 76 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 75.33613613613613 Std Accuracy: 4.477766170241681 Hidden Node mean 4.6967967967967965 Hidden Node std: 2.283697664922948 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
99% (1999 of 2000) |################### | Elapsed Time: 0:00:38 ETA: 0:00:00
=== Performance result === Accuracy: 54.238919459729864 (+/-) 1.5656377754252186 Testing Loss: 0.6961198691847087 (+/-) 0.006240284642032272
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.2941860384159063 Recall: 0.5423891945972986 F1 score: 0.3814679711792393 Testing Time: 0.00744759446087332 (+/-) 0.004837734713018316 Training Time: 4.980670266774012e-07 (+/-) 2.226309467707506e-05 === Average network evolution === Total hidden node: 6.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 6 No. of parameters : 114 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
99% (1999 of 2000) |################### | Elapsed Time: 0:00:37 ETA: 0:00:00
=== Performance result === Accuracy: 56.175337668834416 (+/-) 1.5605879632152218 Testing Loss: 0.6751180544324134 (+/-) 0.00509114181611653 Precision: 0.7132905022374537 Recall: 0.5617533766883441 F1 score: 0.42707030865874057 Testing Time: 0.007155278135741455 (+/-) 0.00470282663310498 Training Time: 1.4965864585124594e-06 (+/-) 3.860303540900367e-05 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 4 No. of parameters : 76 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:39 ETA: 00:00:00
=== Performance result === Accuracy: 54.31745872936469 (+/-) 1.5635833017602094 Testing Loss: 0.6902488746781419 (+/-) 0.006759512549503921 Precision: 0.6900003491437942 Recall: 0.5431745872936469 F1 score: 0.38349040597078354 Testing Time: 0.007570054186410222 (+/-) 0.0049033655642956985 Training Time: 4.996175226895675e-07 (+/-) 2.23324002878993e-05 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 3 No. of parameters : 57 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:38 ETA: 00:00:00
=== Performance result === Accuracy: 54.53011505752877 (+/-) 1.5618300782623784 Testing Loss: 0.6860506683066226 (+/-) 0.0017765962104314317 Precision: 0.6997897003148242 Recall: 0.5453011505752876 F1 score: 0.38874273360211703 Testing Time: 0.007305128088946817 (+/-) 0.004742141532270598 Training Time: 1.4961093828164082e-06 (+/-) 3.859072969178409e-05 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 4 No. of parameters : 76 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (2000 of 2000) |####################| Elapsed Time: 0:00:38 ETA: 00:00:00
=== Performance result === Accuracy: 54.238919459729864 (+/-) 1.5656377754252186 Testing Loss: 0.702019528933559 (+/-) 0.008200431371711172
C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1272: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
Precision: 0.2941860384159063 Recall: 0.5423891945972986 F1 score: 0.3814679711792393 Testing Time: 0.007169294500422514 (+/-) 0.004712738245836063 Training Time: 4.98782640221478e-07 (+/-) 2.229508188207085e-05 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=18, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 18 No. of nodes : 4 No. of parameters : 76 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 54.69925925925926 Std Accuracy: 1.731756458710097 Hidden Node mean 4.2 Hidden Node std: 0.9797958971132714 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [9]:
%run NADINE_classification_electricitypricing.ipynb
Number of input: 8 Number of output: 2 Number of batch: 45 All Data
100% (45 of 45) |########################| Elapsed Time: 0:02:17 ETA: 00:00:00
=== Performance result === Accuracy: 56.088636363636354 (+/-) 7.873436811926857 Testing Loss: 0.6900654421611265 (+/-) 0.021927531906216182 Precision: 0.5292060331246423 Recall: 0.5608863636363637 F1 score: 0.5109524497303286 Testing Time: 0.004230076616460627 (+/-) 0.005702403309514926 Training Time: 3.123543219132857 (+/-) 3.468867672182119 === Average network evolution === Total hidden node: 15.840909090909092 (+/-) 6.349456604588593 Number of layer: 3.5454545454545454 (+/-) 0.9875254992000196 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 3 No. of parameters : 27 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 4 No. of parameters : 16 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 6 No. of parameters : 30 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 7 No. of parameters : 49 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 5 No. of parameters : 40 6 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001]
100% (45 of 45) |########################| Elapsed Time: 0:01:43 ETA: 00:00:00
=== Performance result === Accuracy: 61.5 (+/-) 7.610369719844579 Testing Loss: 0.6423078761859373 (+/-) 0.041741273061774685 Precision: 0.6054952803224337 Recall: 0.615 F1 score: 0.5984579458875032 Testing Time: 0.0024590275504372335 (+/-) 0.0009631354212277176 Training Time: 2.3424539999528364 (+/-) 2.0055707400230265 === Average network evolution === Total hidden node: 9.136363636363637 (+/-) 5.12831227372705 Number of layer: 2.1363636363636362 (+/-) 1.198311484224006 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 4 No. of parameters : 16 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 7 No. of parameters : 35 5 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001]
100% (45 of 45) |########################| Elapsed Time: 0:01:32 ETA: 00:00:00
=== Performance result === Accuracy: 56.929545454545455 (+/-) 6.475498560849987 Testing Loss: 0.6847799542275342 (+/-) 0.026461256559532322 Precision: 0.5414415818550379 Recall: 0.5692954545454545 F1 score: 0.5185883634912459 Testing Time: 0.002593230117451061 (+/-) 0.0007136240333021079 Training Time: 2.087898514487527 (+/-) 1.230420982387765 === Average network evolution === Total hidden node: 9.727272727272727 (+/-) 1.8508430065002117 Number of layer: 2.159090909090909 (+/-) 0.6008774575637402 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=2, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 2 No. of parameters : 16 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=2, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 2 No. of nodes : 4 No. of parameters : 12 4 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001]
100% (45 of 45) |########################| Elapsed Time: 0:01:51 ETA: 00:00:00
=== Performance result === Accuracy: 58.41136363636363 (+/-) 6.74985575297357 Testing Loss: 0.6794444512237202 (+/-) 0.026962036030643524 Precision: 0.5659518657319237 Recall: 0.5841136363636363 F1 score: 0.5489847390526132 Testing Time: 0.002870180390097878 (+/-) 0.0008821189022379835 Training Time: 2.5197523073716597 (+/-) 2.299057321611756 === Average network evolution === Total hidden node: 11.431818181818182 (+/-) 5.210992434149212 Number of layer: 2.5454545454545454 (+/-) 0.9875254992000196 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 4 No. of parameters : 24 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 8 No. of parameters : 32 5 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001]
100% (45 of 45) |########################| Elapsed Time: 0:01:13 ETA: 00:00:00
=== Performance result === Accuracy: 61.945454545454545 (+/-) 7.774213042472664 Testing Loss: 0.6325082670558583 (+/-) 0.05834837343200157 Precision: 0.6111065420140294 Recall: 0.6194545454545455 F1 score: 0.607888904184372 Testing Time: 0.002092578194358132 (+/-) 0.0006768741018981584 Training Time: 1.6607387878678062 (+/-) 0.5026806543196276 === Average network evolution === Total hidden node: 7.704545454545454 (+/-) 3.0568815471634534 Number of layer: 1.25 (+/-) 0.4330127018922193 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 6 No. of parameters : 48 3 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001]
Mean Accuracy: 59.02976744186047 Std Accuracy: 7.511024297153375 Hidden Node mean 10.874418604651163 Hidden Node std: 5.416430345276791 Hidden Layer mean: 2.353488372093023 Hidden Layer std: 1.1556431486726748 50% Data
100% (45 of 45) |########################| Elapsed Time: 0:00:33 ETA: 00:00:00
=== Performance result === Accuracy: 62.69318181818183 (+/-) 6.89648783233086 Testing Loss: 0.6258891265500676 (+/-) 0.04981666813935529 Precision: 0.6203247289864602 Recall: 0.6269318181818182 F1 score: 0.6197309257123984 Testing Time: 0.0017736229029568758 (+/-) 0.000455783268869294 Training Time: 0.7680414048108187 (+/-) 0.0127041853020773 === Average network evolution === Total hidden node: 5.9772727272727275 (+/-) 0.14903269373413638 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:33 ETA: 00:00:00
=== Performance result === Accuracy: 60.75454545454544 (+/-) 7.98602343146662 Testing Loss: 0.6463055244900964 (+/-) 0.05594611827292049 Precision: 0.598994370683884 Recall: 0.6075454545454545 F1 score: 0.5980861796639305 Testing Time: 0.0018613934516906738 (+/-) 0.00048738213544072256 Training Time: 0.7625350085171786 (+/-) 0.016469567349510723 === Average network evolution === Total hidden node: 6.545454545454546 (+/-) 0.49792959773196915 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:33 ETA: 00:00:00
=== Performance result === Accuracy: 64.00227272727273 (+/-) 7.8633640914503475 Testing Loss: 0.6203348223458637 (+/-) 0.059221981054932225 Precision: 0.6363734525591423 Recall: 0.6400227272727272 F1 score: 0.6372628837230548 Testing Time: 0.0017918727614662864 (+/-) 0.00038235176300413146 Training Time: 0.7649643204428933 (+/-) 0.013774717020110839 === Average network evolution === Total hidden node: 5.545454545454546 (+/-) 0.49792959773196915 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:33 ETA: 00:00:00
=== Performance result === Accuracy: 61.415909090909096 (+/-) 6.593727845523081 Testing Loss: 0.6425982469862158 (+/-) 0.04851161347292417 Precision: 0.6044252337949235 Recall: 0.6141590909090909 F1 score: 0.5933801247284796 Testing Time: 0.0017538612539117987 (+/-) 0.00041422620359300206 Training Time: 0.7647274082357233 (+/-) 0.011661524538827595 === Average network evolution === Total hidden node: 5.545454545454546 (+/-) 0.49792959773196915 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:33 ETA: 00:00:00
=== Performance result === Accuracy: 60.57727272727273 (+/-) 7.214816309523431 Testing Loss: 0.6459532434290106 (+/-) 0.054744189003492226 Precision: 0.5944299435314258 Recall: 0.6057727272727272 F1 score: 0.5838548310389271 Testing Time: 0.0018204342235218394 (+/-) 0.0004188041145837858 Training Time: 0.7599805918606845 (+/-) 0.011473545585987729 === Average network evolution === Total hidden node: 7.045454545454546 (+/-) 0.20829889522526543 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 62.26883720930232 Std Accuracy: 6.701010409511607 Hidden Node mean 6.1395348837209305 Hidden Node std: 0.7081385802109118 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 25% Data
100% (45 of 45) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 61.70454545454545 (+/-) 6.525542629664621 Testing Loss: 0.6409691951491616 (+/-) 0.03997752527110094 Precision: 0.6077891832596168 Recall: 0.6170454545454546 F1 score: 0.5995189252476162 Testing Time: 0.0018435554070906205 (+/-) 0.0004977706414413864 Training Time: 0.40245831554586237 (+/-) 0.011997636735961159 === Average network evolution === Total hidden node: 4.7727272727272725 (+/-) 0.4701854742176637 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 58.80227272727273 (+/-) 6.416438003510402 Testing Loss: 0.6626005118543451 (+/-) 0.035105224101239625 Precision: 0.5718388536749407 Recall: 0.5880227272727273 F1 score: 0.5581361538698065 Testing Time: 0.0016653266820040617 (+/-) 0.000508609999709917 Training Time: 0.4031171744520014 (+/-) 0.013154746109783004 === Average network evolution === Total hidden node: 4.545454545454546 (+/-) 0.49792959773196915 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 61.775 (+/-) 6.897039055078109 Testing Loss: 0.6418313722718846 (+/-) 0.044070089256331144 Precision: 0.6089288449330501 Recall: 0.61775 F1 score: 0.6044201063818361 Testing Time: 0.0019264492121609774 (+/-) 0.0004233096868348012 Training Time: 0.39600146358663385 (+/-) 0.009266562601823224 === Average network evolution === Total hidden node: 6.590909090909091 (+/-) 0.4916660830178168 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 58.963636363636354 (+/-) 6.457451330513488 Testing Loss: 0.6619926961985502 (+/-) 0.03642827155706012 Precision: 0.5737830731039351 Recall: 0.5896363636363636 F1 score: 0.55902165301373 Testing Time: 0.0017501657659357245 (+/-) 0.00041480783674129913 Training Time: 0.39625219865278766 (+/-) 0.008553493608402072 === Average network evolution === Total hidden node: 4.5 (+/-) 0.5 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 59.2 (+/-) 7.5700966728545565 Testing Loss: 0.6524218361486088 (+/-) 0.038175686018998296 Precision: 0.5787635492117729 Recall: 0.592 F1 score: 0.5727082914759181 Testing Time: 0.001841138709675182 (+/-) 0.0005011335921089303 Training Time: 0.3974649581042203 (+/-) 0.013749878369493908 === Average network evolution === Total hidden node: 5.4772727272727275 (+/-) 0.6211659258507477 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 60.153488372093015 Std Accuracy: 6.5577062367167755 Hidden Node mean 5.176744186046512 Hidden Node std: 0.938290679733898 Hidden Layer mean: 1.0 Hidden Layer std: 0.0 Infinite Delay
88% (40 of 45) |##################### | Elapsed Time: 0:00:00 ETA: 00:00:00
=== Performance result === Accuracy: 57.4840909090909 (+/-) 6.357384510151746 Testing Loss: 0.6895422718741677 (+/-) 0.006026430354996066 Precision: 0.5449361382569123 Recall: 0.5748409090909091 F1 score: 0.5002452441554814 Testing Time: 0.0009053349494934082 (+/-) 0.00041547511089249164 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 5 No. of parameters : 45 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
100% (45 of 45) |########################| Elapsed Time: 0:00:00 ETA: 00:00:00
=== Performance result === Accuracy: 43.56363636363636 (+/-) 7.04392358732917 Testing Loss: 0.7470100210471586 (+/-) 0.04702806080860336 Precision: 0.6122641980725461 Recall: 0.43563636363636365 F1 score: 0.290055142646441 Testing Time: 0.0008377703753384677 (+/-) 0.0004206037549611421 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 3 No. of parameters : 27 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
95% (43 of 45) |###################### | Elapsed Time: 0:00:00 ETA: 00:00:00
=== Performance result === Accuracy: 57.75909090909091 (+/-) 6.413420523396026 Testing Loss: 0.6821321939880197 (+/-) 0.02238828870737873 Precision: 0.48578461625923824 Recall: 0.577590909090909 F1 score: 0.42342251171979395 Testing Time: 0.0008845546028830788 (+/-) 0.0003815061135243028 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
86% (39 of 45) |#################### | Elapsed Time: 0:00:00 ETA: 00:00:00
=== Performance result === Accuracy: 52.24772727272725 (+/-) 5.4648250673309695 Testing Loss: 0.6910624761473049 (+/-) 0.005688135580340763 Precision: 0.5263060247681588 Recall: 0.5224772727272727 F1 score: 0.5240679518098668 Testing Time: 0.0009294314817948775 (+/-) 0.0005795397471903198 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 4 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 45) | | Elapsed Time: 0:00:00 ETA: --:--:--
=== Performance result === Accuracy: 48.13863636363636 (+/-) 7.012528266320958 Testing Loss: 0.6980154256929051 (+/-) 0.014251611261498246 Precision: 0.5814207155948179 Recall: 0.4813863636363636 F1 score: 0.42104715934853 Testing Time: 0.0007011944597417659 (+/-) 0.0005004274457939448 Training Time: 2.2666020826859906e-05 (+/-) 0.00014863103816268263 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 3 No. of parameters : 27 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
Mean Accuracy: 51.712093023255804 Std Accuracy: 8.201408999235559 Hidden Node mean 3.8 Hidden Node std: 0.7483314773547883 Hidden Layer mean: 1.0 Hidden Layer std: 0.0
In [1]:
%run NADINE_classification_occupancy.ipynb
Number of input: 5 Number of output: 2 Number of batch: 20 All Data
100% (20 of 20) |########################| Elapsed Time: 0:01:54 ETA: 00:00:00
=== Performance result === Accuracy: 66.11578947368422 (+/-) 30.047370725816027 Testing Loss: 0.7413052906723399 (+/-) 0.6841498508550296 Precision: 0.6181622192982457 Recall: 0.6611578947368421 F1 score: 0.6374759268636577 Testing Time: 0.0067740239595112045 (+/-) 0.007969442062941503 Training Time: 6.031864680741963 (+/-) 4.475899293238954 === Average network evolution === Total hidden node: 37.578947368421055 (+/-) 16.432688185358778 Number of layer: 6.2631578947368425 (+/-) 2.935124946020776 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 4 No. of parameters : 32 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 9 No. of parameters : 45 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=9, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 9 No. of nodes : 7 No. of parameters : 70 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 6 No. of parameters : 48 6 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 5 No. of parameters : 35 7 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 8 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 4 No. of parameters : 32 9 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 10 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 4 No. of parameters : 24 11 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 12 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.02, 0.02]
100% (20 of 20) |########################| Elapsed Time: 0:01:39 ETA: 00:00:00
=== Performance result === Accuracy: 71.83157894736841 (+/-) 26.190557813066082 Testing Loss: 0.7240843404893225 (+/-) 0.7491465573109074 Precision: 0.5845494269005848 Recall: 0.7183157894736842 F1 score: 0.6445656508337739 Testing Time: 0.008803857000250565 (+/-) 0.010305741310137373 Training Time: 5.204964411886115 (+/-) 3.117274289647815 === Average network evolution === Total hidden node: 45.73684210526316 (+/-) 19.401145124269444 Number of layer: 6.157894736842105 (+/-) 2.7959210495354583 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 7 No. of parameters : 56 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 12 No. of parameters : 96 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=12, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 12 No. of nodes : 5 No. of parameters : 65 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 8 No. of parameters : 48 6 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 7 No. of parameters : 63 7 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 6 No. of parameters : 48 8 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 6 No. of parameters : 42 9 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 9 No. of parameters : 63 10 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=9, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 9 No. of nodes : 6 No. of parameters : 60 11 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.02]
100% (20 of 20) |########################| Elapsed Time: 0:01:52 ETA: 00:00:00
=== Performance result === Accuracy: 68.9578947368421 (+/-) 28.34010206476948 Testing Loss: 0.7092076442449501 (+/-) 0.6989813017572891 Precision: 0.6098767306501547 Recall: 0.6895789473684211 F1 score: 0.6436429043085233 Testing Time: 0.006817290657445004 (+/-) 0.009348207849188047 Training Time: 5.91120171546936 (+/-) 3.8944360573062093 === Average network evolution === Total hidden node: 35.73684210526316 (+/-) 15.592987983007848 Number of layer: 6.2631578947368425 (+/-) 2.935124946020776 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 5 No. of parameters : 30 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 5 No. of parameters : 30 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 10 No. of parameters : 60 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=10, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 10 No. of nodes : 6 No. of parameters : 66 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 6 No. of parameters : 42 6 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 4 No. of parameters : 28 7 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 8 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 4 No. of parameters : 24 9 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 4 No. of parameters : 20 10 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 7 No. of parameters : 35 11 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 7 No. of parameters : 56 12 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.02, 0.02]
100% (20 of 20) |########################| Elapsed Time: 0:01:42 ETA: 00:00:00
=== Performance result === Accuracy: 66.0263157894737 (+/-) 29.97484633392442 Testing Loss: 0.7211671401746571 (+/-) 0.6458575280478268 Precision: 0.633785830406839 Recall: 0.6602631578947369 F1 score: 0.6459800711196622 Testing Time: 0.005980027349371659 (+/-) 0.009533765487810101 Training Time: 5.392461889668515 (+/-) 4.421539000421144 === Average network evolution === Total hidden node: 28.36842105263158 (+/-) 14.970885873010811 Number of layer: 5.0 (+/-) 2.809757434745082 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 8 No. of parameters : 48 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 4 No. of parameters : 28 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 5 No. of parameters : 25 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 6 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 7 No. of parameters : 56 7 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 3 No. of parameters : 24 8 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 5 No. of parameters : 20 9 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 4 No. of parameters : 24 10 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 3 No. of parameters : 15 11 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.02, 0.02]
100% (20 of 20) |########################| Elapsed Time: 0:01:45 ETA: 00:00:00
=== Performance result === Accuracy: 72.44210526315791 (+/-) 23.262604184884456 Testing Loss: 0.730884123299467 (+/-) 0.7203345006255195 Precision: 0.6297510687158236 Recall: 0.724421052631579 F1 score: 0.6642873264134467 Testing Time: 0.007869143235056024 (+/-) 0.009911177341382168 Training Time: 5.54103392048886 (+/-) 3.8335136247810304 === Average network evolution === Total hidden node: 45.78947368421053 (+/-) 19.294337722348605 Number of layer: 6.2631578947368425 (+/-) 2.935124946020776 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 9 No. of parameters : 54 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=9, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 9 No. of nodes : 4 No. of parameters : 40 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 13 No. of parameters : 65 4 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=13, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 13 No. of nodes : 6 No. of parameters : 84 5 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 8 No. of parameters : 56 6 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=8, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 8 No. of nodes : 6 No. of parameters : 54 7 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 7 No. of parameters : 49 8 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 7 No. of parameters : 56 9 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 5 No. of parameters : 40 10 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 3 No. of parameters : 18 11 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=3, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 3 No. of nodes : 4 No. of parameters : 16 12 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.0001, 0.02, 0.02]
========== Performance occupancy ========== Preq Accuracy: 69.07 (+/-) 2.72 F1 score: 0.65 (+/-) 0.01 Precision: 0.62 (+/-) 0.02 Recall: 0.69 (+/-) 0.03 Training time: 5.62 (+/-) 0.31 Testing time: 0.01 (+/-) 0.0 ========== Network ========== Number of hidden layers: 10.6 (+/-) 0.49 Number of features: 64.2 (+/-) 7.73 50% Data
100% (20 of 20) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 83.76842105263158 (+/-) 18.805963573504492 Testing Loss: 0.3486570302200945 (+/-) 0.35759483750829896 Precision: 0.827128067000654 Recall: 0.8376842105263158 F1 score: 0.8270684206800012 Testing Time: 0.00236117212395919 (+/-) 0.00048536545356259246 Training Time: 0.9028193574202689 (+/-) 0.061089705891232776 === Average network evolution === Total hidden node: 10.31578947368421 (+/-) 2.3181805837416043 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 14 No. of parameters : 84 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=14, out_features=2, bias=True) ) No. of inputs : 14 No. of output : 2 No. of parameters : 30 Dynamic laerning rate for each hidden layer: [0.02]
100% (20 of 20) |########################| Elapsed Time: 0:00:24 ETA: 00:00:00
=== Performance result === Accuracy: 62.5263157894737 (+/-) 32.48698269032275 Testing Loss: 0.7588729304996761 (+/-) 0.5900879435789674 Precision: 0.6167312729617602 Recall: 0.6252631578947369 F1 score: 0.6209093120939793 Testing Time: 0.0030910592330129524 (+/-) 0.0011165618252967848 Training Time: 1.269457177111977 (+/-) 0.4350584521434636 === Average network evolution === Total hidden node: 18.157894736842106 (+/-) 5.460510462881237 Number of layer: 2.210526315789474 (+/-) 0.6137844099837159 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=7, out_features=11, bias=True) (activation): Sigmoid() ) No. of inputs : 7 No. of nodes : 11 No. of parameters : 88 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=11, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 11 No. of nodes : 7 No. of parameters : 84 4 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001]
100% (20 of 20) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 81.10526315789475 (+/-) 18.33180938852092 Testing Loss: 0.4811954941894663 (+/-) 0.4929174435311406 Precision: 0.7952554222634699 Recall: 0.8110526315789474 F1 score: 0.7974025203218573 Testing Time: 0.0020365087609542044 (+/-) 0.0005069820694998896 Training Time: 0.9019824705625835 (+/-) 0.01403862773792278 === Average network evolution === Total hidden node: 7.7894736842105265 (+/-) 1.9078493642517746 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=11, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 11 No. of parameters : 66 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=11, out_features=2, bias=True) ) No. of inputs : 11 No. of output : 2 No. of parameters : 24 Dynamic laerning rate for each hidden layer: [0.02]
100% (20 of 20) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 77.54736842105265 (+/-) 21.379635377252484 Testing Loss: 0.5326100588335019 (+/-) 0.5236876285111566 Precision: 0.753117093742324 Recall: 0.7754736842105263 F1 score: 0.7594887861467363 Testing Time: 0.002092072838231137 (+/-) 0.0006377409806986178 Training Time: 0.9000320936504164 (+/-) 0.01737350351336918 === Average network evolution === Total hidden node: 9.894736842105264 (+/-) 1.483052926695302 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=13, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 13 No. of parameters : 78 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=13, out_features=2, bias=True) ) No. of inputs : 13 No. of output : 2 No. of parameters : 28 Dynamic laerning rate for each hidden layer: [0.02]
100% (20 of 20) |########################| Elapsed Time: 0:00:17 ETA: 00:00:00
=== Performance result === Accuracy: 84.27894736842104 (+/-) 18.43767821173087 Testing Loss: 0.431255930323938 (+/-) 0.4444738051287629 Precision: 0.8337756993794551 Recall: 0.8427894736842105 F1 score: 0.8283091559219236 Testing Time: 0.0019918366482383327 (+/-) 0.00045626979282153117 Training Time: 0.9034348663530851 (+/-) 0.012302593071830644 === Average network evolution === Total hidden node: 6.473684210526316 (+/-) 1.6015920057582043 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 9 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance occupancy ========== Preq Accuracy: 77.85 (+/-) 8.02 F1 score: 0.77 (+/-) 0.08 Precision: 0.77 (+/-) 0.08 Recall: 0.78 (+/-) 0.08 Training time: 0.98 (+/-) 0.15 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.4 (+/-) 0.8 Number of features: 14.4 (+/-) 5.57 25% Data
100% (20 of 20) |########################| Elapsed Time: 0:00:10 ETA: 00:00:00
=== Performance result === Accuracy: 66.11578947368422 (+/-) 30.047370725816027 Testing Loss: 0.7770996762831744 (+/-) 0.6202127485966321 Precision: 0.6181622192982457 Recall: 0.6611578947368421 F1 score: 0.6374759268636577 Testing Time: 0.002880121532239412 (+/-) 0.0008516272331530303 Training Time: 0.5707024775053325 (+/-) 0.06804353319753645 === Average network evolution === Total hidden node: 15.578947368421053 (+/-) 3.7740915877889587 Number of layer: 1.894736842105263 (+/-) 0.30689220499185793 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 6 No. of parameters : 36 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 12 No. of parameters : 84 3 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001]
100% (20 of 20) |########################| Elapsed Time: 0:00:12 ETA: 00:00:00
=== Performance result === Accuracy: 61.78947368421054 (+/-) 33.073170789006326 Testing Loss: 0.7050064955102769 (+/-) 0.5140652787407359 Precision: 0.6124252984109287 Recall: 0.6178947368421053 F1 score: 0.6151255657920228 Testing Time: 0.0034108914827045644 (+/-) 0.0008668975462704134 Training Time: 0.6402965721331144 (+/-) 0.21095012193711943 === Average network evolution === Total hidden node: 17.736842105263158 (+/-) 5.692489098185783 Number of layer: 2.210526315789474 (+/-) 0.6137844099837159 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 4 No. of parameters : 24 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=4, out_features=14, bias=True) (activation): Sigmoid() ) No. of inputs : 4 No. of nodes : 14 No. of parameters : 70 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=14, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 14 No. of nodes : 7 No. of parameters : 105 4 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001]
100% (20 of 20) |########################| Elapsed Time: 0:00:10 ETA: 00:00:00
=== Performance result === Accuracy: 64.86842105263159 (+/-) 29.091562398229545 Testing Loss: 0.7032000747950453 (+/-) 0.5998521724690425 Precision: 0.6127857725625887 Recall: 0.6486842105263158 F1 score: 0.6293163794854115 Testing Time: 0.0028353239360608554 (+/-) 0.0006679077558093613 Training Time: 0.5672339012748316 (+/-) 0.07621184706673266 === Average network evolution === Total hidden node: 14.842105263157896 (+/-) 3.616674219496732 Number of layer: 1.894736842105263 (+/-) 0.30689220499185793 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 5 No. of parameters : 30 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 12 No. of parameters : 72 3 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001]
100% (20 of 20) |########################| Elapsed Time: 0:00:12 ETA: 00:00:00
=== Performance result === Accuracy: 60.852631578947374 (+/-) 32.32311163403627 Testing Loss: 0.7280160220045793 (+/-) 0.5717919671517209 Precision: 0.5969272982456141 Recall: 0.6085263157894737 F1 score: 0.6026007831949999 Testing Time: 0.0033537463137978 (+/-) 0.0012572198346441784 Training Time: 0.6365938939546284 (+/-) 0.20844465429570497 === Average network evolution === Total hidden node: 18.63157894736842 (+/-) 6.276191657163722 Number of layer: 2.210526315789474 (+/-) 0.6137844099837159 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 6 No. of parameters : 36 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 12 No. of parameters : 84 3 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=12, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 12 No. of nodes : 9 No. of parameters : 117 4 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001, 0.0001]
100% (20 of 20) |########################| Elapsed Time: 0:00:10 ETA: 00:00:00
=== Performance result === Accuracy: 66.11578947368422 (+/-) 30.047370725816027 Testing Loss: 0.7305813410172337 (+/-) 0.6391395071522439 Precision: 0.6181622192982457 Recall: 0.6611578947368421 F1 score: 0.6374759268636577 Testing Time: 0.0029271778307462994 (+/-) 0.0006789698897705654 Training Time: 0.5708209840874923 (+/-) 0.057423669328512034 === Average network evolution === Total hidden node: 15.631578947368421 (+/-) 3.730535820351442 Number of layer: 1.894736842105263 (+/-) 0.30689220499185793 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 6 No. of parameters : 36 2 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=6, out_features=12, bias=True) (activation): Sigmoid() ) No. of inputs : 6 No. of nodes : 12 No. of parameters : 84 3 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=12, out_features=2, bias=True) ) No. of inputs : 12 No. of output : 2 No. of parameters : 26 Dynamic laerning rate for each hidden layer: [0.0001, 0.0001]
========== Performance occupancy ========== Preq Accuracy: 63.95 (+/-) 2.21 F1 score: 0.62 (+/-) 0.01 Precision: 0.61 (+/-) 0.01 Recall: 0.64 (+/-) 0.02 Training time: 0.6 (+/-) 0.03 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 2.4 (+/-) 0.49 Number of features: 21.0 (+/-) 4.15 Infinite Delay
N/A% (0 of 20) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.09473684210526 (+/-) 20.698753425068876 Testing Loss: 0.5331742975272631 (+/-) 0.30023566023729126 Precision: 0.5943598448753463 Recall: 0.7709473684210526 F1 score: 0.6712337763095327 Testing Time: 0.0015766746119449014 (+/-) 0.0005872146228732684 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 10.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=10, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 10 No. of parameters : 60 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=10, out_features=2, bias=True) ) No. of inputs : 10 No. of output : 2 No. of parameters : 22 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 20) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.09473684210526 (+/-) 20.698753425068876 Testing Loss: 0.5431874182663465 (+/-) 0.3214055440872702 Precision: 0.5943598448753463 Recall: 0.7709473684210526 F1 score: 0.6712337763095327 Testing Time: 0.0014715194702148438 (+/-) 0.0005958592622866412 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 3.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=3, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 3 No. of parameters : 18 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=3, out_features=2, bias=True) ) No. of inputs : 3 No. of output : 2 No. of parameters : 8 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 20) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.09473684210526 (+/-) 20.698753425068876 Testing Loss: 0.5280480431882959 (+/-) 0.26261602505832626 Precision: 0.5943598448753463 Recall: 0.7709473684210526 F1 score: 0.6712337763095327 Testing Time: 0.0016278091229890521 (+/-) 0.0005813392528768669 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 7 No. of parameters : 42 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 20) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.09473684210526 (+/-) 20.698753425068876 Testing Loss: 0.5729194726598891 (+/-) 0.38271157369239533 Precision: 0.5943598448753463 Recall: 0.7709473684210526 F1 score: 0.6712337763095327 Testing Time: 0.001837215925517835 (+/-) 0.0007423172491265569 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 9.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 9 No. of parameters : 54 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 20) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.09473684210526 (+/-) 20.698753425068876 Testing Loss: 0.4863303711539821 (+/-) 0.2862979696687378 Precision: 0.5943598448753463 Recall: 0.7709473684210526 F1 score: 0.6712337763095327 Testing Time: 0.0014715320185611123 (+/-) 0.000678633384282552 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 6.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=5, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 5 No. of nodes : 6 No. of parameters : 36 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance occupancy ========== Preq Accuracy: 77.09 (+/-) 0.0 F1 score: 0.67 (+/-) 0.0 Precision: 0.59 (+/-) 0.0 Recall: 0.77 (+/-) 0.0 Training time: 0.0 (+/-) 0.0 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.0 (+/-) 0.0 Number of features: 7.0 (+/-) 2.45
In [2]:
%run NADINE_classification_creditcarddefault.ipynb
Number of input: 24 Number of output: 2 Number of batch: 30 All Data
100% (30 of 30) |########################| Elapsed Time: 0:00:53 ETA: 00:00:00
=== Performance result === Accuracy: 79.70344827586207 (+/-) 2.6260607391957453 Testing Loss: 0.4876625393999034 (+/-) 0.034046624501000204 Precision: 0.779108400805486 Recall: 0.7970344827586207 F1 score: 0.7409359345760664 Testing Time: 0.0022601423592403017 (+/-) 0.0005126260840128597 Training Time: 1.8320883635816902 (+/-) 0.17969217116094482 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:48 ETA: 00:00:00
=== Performance result === Accuracy: 79.40689655172413 (+/-) 2.3881793721235414 Testing Loss: 0.48319352495259255 (+/-) 0.03468998446802365 Precision: 0.7785146352377841 Recall: 0.7940689655172414 F1 score: 0.7311199269812312 Testing Time: 0.0021869725194470636 (+/-) 0.0006640042668574176 Training Time: 1.6570927439064815 (+/-) 0.07569471133639247 === Average network evolution === Total hidden node: 7.413793103448276 (+/-) 0.49251230541674823 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:49 ETA: 00:00:00
=== Performance result === Accuracy: 79.2655172413793 (+/-) 2.774051669218841 Testing Loss: 0.488578534331815 (+/-) 0.03856925654889938 Precision: 0.7756216598111582 Recall: 0.7926551724137931 F1 score: 0.728010691098317 Testing Time: 0.002055003725249192 (+/-) 0.0004413418572183541 Training Time: 1.692105169953971 (+/-) 0.07951530974987778 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:48 ETA: 00:00:00
=== Performance result === Accuracy: 79.95172413793104 (+/-) 2.4609530939854727 Testing Loss: 0.48045521152430565 (+/-) 0.03314798795540664 Precision: 0.7794362771483293 Recall: 0.7995172413793104 F1 score: 0.7498015398817669 Testing Time: 0.00240176299522663 (+/-) 0.000614933402361267 Training Time: 1.6598528500260978 (+/-) 0.05700683132292424 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:53 ETA: 00:00:00
=== Performance result === Accuracy: 79.37931034482759 (+/-) 2.6367926324314372 Testing Loss: 0.4852180213763796 (+/-) 0.036050494308274665 Precision: 0.7739958725871223 Recall: 0.7937931034482759 F1 score: 0.7332104440612969 Testing Time: 0.0021968545584843077 (+/-) 0.0005484229798818605 Training Time: 1.8407955991810765 (+/-) 0.12543995829021518 === Average network evolution === Total hidden node: 7.413793103448276 (+/-) 0.49251230541674823 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance creditcarddefault ========== Preq Accuracy: 79.54 (+/-) 0.25 F1 score: 0.74 (+/-) 0.01 Precision: 0.78 (+/-) 0.0 Recall: 0.8 (+/-) 0.0 Training time: 1.74 (+/-) 0.08 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.0 (+/-) 0.0 Number of features: 7.6 (+/-) 0.49 50% Data
100% (30 of 30) |########################| Elapsed Time: 0:00:26 ETA: 00:00:00
=== Performance result === Accuracy: 78.71379310344827 (+/-) 2.6797740689547123 Testing Loss: 0.49564206086356066 (+/-) 0.04097241479918927 Precision: 0.7810204599131659 Recall: 0.7871379310344827 F1 score: 0.7079674062215365 Testing Time: 0.002222949060900458 (+/-) 0.0005032427139756378 Training Time: 0.8979875548132534 (+/-) 0.05336606353053322 === Average network evolution === Total hidden node: 8.448275862068966 (+/-) 0.49731741730537776 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 9 No. of parameters : 225 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:25 ETA: 00:00:00
=== Performance result === Accuracy: 78.72068965517242 (+/-) 2.6511388977966095 Testing Loss: 0.4994287048948222 (+/-) 0.037873722835753916 Precision: 0.7722776146913933 Recall: 0.7872068965517242 F1 score: 0.7107548640785002 Testing Time: 0.002148052741741312 (+/-) 0.0003801168121774236 Training Time: 0.8738716552997458 (+/-) 0.04450860471135495 === Average network evolution === Total hidden node: 8.724137931034482 (+/-) 0.4469476343729558 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 9 No. of parameters : 225 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:26 ETA: 00:00:00
=== Performance result === Accuracy: 78.81724137931036 (+/-) 2.9047011794220805 Testing Loss: 0.4969960122272886 (+/-) 0.03326975250758636 Precision: 0.7688784487331931 Recall: 0.7881724137931034 F1 score: 0.7156120515933921 Testing Time: 0.0018695140707081761 (+/-) 0.0006088282308082159 Training Time: 0.9034510563159811 (+/-) 0.05763563938445578 === Average network evolution === Total hidden node: 4.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=4, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 4 No. of parameters : 100 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=4, out_features=2, bias=True) ) No. of inputs : 4 No. of output : 2 No. of parameters : 10 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:25 ETA: 00:00:00
=== Performance result === Accuracy: 78.83793103448275 (+/-) 2.6816636361245467 Testing Loss: 0.4930659645590289 (+/-) 0.03484018083557293 Precision: 0.7702990128734366 Recall: 0.7883793103448276 F1 score: 0.7156915108240074 Testing Time: 0.0020897799524767645 (+/-) 0.0007124881861494125 Training Time: 0.8853979768424198 (+/-) 0.062073183108780176 === Average network evolution === Total hidden node: 7.586206896551724 (+/-) 0.49251230541674823 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:27 ETA: 00:00:00
=== Performance result === Accuracy: 79.1655172413793 (+/-) 2.671209962450695 Testing Loss: 0.49398861465782956 (+/-) 0.03174511796041849 Precision: 0.7700824513352108 Recall: 0.7916551724137931 F1 score: 0.7280565683385404 Testing Time: 0.002230586676762022 (+/-) 0.0004277909400775066 Training Time: 0.9384993602489603 (+/-) 0.09489851387399385 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance creditcarddefault ========== Preq Accuracy: 78.85 (+/-) 0.16 F1 score: 0.72 (+/-) 0.01 Precision: 0.77 (+/-) 0.0 Recall: 0.79 (+/-) 0.0 Training time: 0.9 (+/-) 0.02 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.0 (+/-) 0.0 Number of features: 7.4 (+/-) 1.85 25% Data
100% (30 of 30) |########################| Elapsed Time: 0:00:13 ETA: 00:00:00
=== Performance result === Accuracy: 77.97241379310346 (+/-) 2.474824487224977 Testing Loss: 0.5105584376844866 (+/-) 0.038857379659534445 Precision: 0.776626186625895 Recall: 0.7797241379310345 F1 score: 0.6853532857222552 Testing Time: 0.002296086015372441 (+/-) 0.000532624951457563 Training Time: 0.47777397879238787 (+/-) 0.02501534995768557 === Average network evolution === Total hidden node: 6.793103448275862 (+/-) 0.40508069394726653 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:13 ETA: 00:00:00
=== Performance result === Accuracy: 78.21034482758623 (+/-) 2.370849524390114 Testing Loss: 0.504766924627896 (+/-) 0.03904537430845272 Precision: 0.7738992890157953 Recall: 0.7821034482758621 F1 score: 0.6932599962246041 Testing Time: 0.0021222870925377154 (+/-) 0.0005105055696071864 Training Time: 0.45900690966639024 (+/-) 0.03577791302815684 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:12 ETA: 00:00:00
=== Performance result === Accuracy: 78.28620689655172 (+/-) 2.556091323436145 Testing Loss: 0.5013325985135704 (+/-) 0.032063591881526694 Precision: 0.7731704937019902 Recall: 0.7828620689655172 F1 score: 0.6958665174243298 Testing Time: 0.002227709211152175 (+/-) 0.0004250723076319963 Training Time: 0.4289821345230629 (+/-) 0.017447438929185703 === Average network evolution === Total hidden node: 9.068965517241379 (+/-) 0.2533954906327425 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=9, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 9 No. of parameters : 225 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=9, out_features=2, bias=True) ) No. of inputs : 9 No. of output : 2 No. of parameters : 20 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:12 ETA: 00:00:00
=== Performance result === Accuracy: 78.13103448275862 (+/-) 2.3799599313167032 Testing Loss: 0.5030371567298626 (+/-) 0.03631214703609974 Precision: 0.7766233416615541 Recall: 0.7813103448275862 F1 score: 0.690414788367686 Testing Time: 0.0019914035139412716 (+/-) 0.00038310826711397407 Training Time: 0.4396825001157563 (+/-) 0.0386737910206393 === Average network evolution === Total hidden node: 7.206896551724138 (+/-) 0.40508069394726653 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
100% (30 of 30) |########################| Elapsed Time: 0:00:13 ETA: 00:00:00
=== Performance result === Accuracy: 77.99310344827585 (+/-) 2.482341919053771 Testing Loss: 0.5059115115938515 (+/-) 0.032325487245102744 Precision: 0.7778109165535297 Recall: 0.7799310344827586 F1 score: 0.6859655269971857 Testing Time: 0.0022628553982438713 (+/-) 0.0005836023150720172 Training Time: 0.4661414623260498 (+/-) 0.0409321808077169 === Average network evolution === Total hidden node: 6.379310344827586 (+/-) 0.48521542343000995 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=6, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 6 No. of parameters : 150 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=6, out_features=2, bias=True) ) No. of inputs : 6 No. of output : 2 No. of parameters : 14 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance creditcarddefault ========== Preq Accuracy: 78.12 (+/-) 0.12 F1 score: 0.69 (+/-) 0.0 Precision: 0.78 (+/-) 0.0 Recall: 0.78 (+/-) 0.0 Training time: 0.45 (+/-) 0.02 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.0 (+/-) 0.0 Number of features: 7.4 (+/-) 1.02 Infinite Delay
N/A% (0 of 30) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.85517241379311 (+/-) 2.505247762115322 Testing Loss: 0.5292846241901661 (+/-) 0.030651198693608425 Precision: 0.6061427871581452 Recall: 0.7785517241379311 F1 score: 0.6816138984678044 Testing Time: 0.0011349710924872037 (+/-) 0.0005689360435162017 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 5.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=5, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 5 No. of parameters : 125 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=5, out_features=2, bias=True) ) No. of inputs : 5 No. of output : 2 No. of parameters : 12 Dynamic laerning rate for each hidden layer: [0.02]
90% (27 of 30) |##################### | Elapsed Time: 0:00:00 ETA: 00:00:00C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.85517241379311 (+/-) 2.505247762115322 Testing Loss: 0.5244946222880791 (+/-) 0.028016860389621246 Precision: 0.6061427871581452 Recall: 0.7785517241379311 F1 score: 0.6816138984678044 Testing Time: 0.0016700481546336207 (+/-) 0.0007442008502926363 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 11.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=11, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 11 No. of parameters : 275 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=11, out_features=2, bias=True) ) No. of inputs : 11 No. of output : 2 No. of parameters : 24 Dynamic laerning rate for each hidden layer: [0.02]
N/A% (0 of 30) | | Elapsed Time: 0:00:00 ETA: --:--:--C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.85517241379311 (+/-) 2.505247762115322 Testing Loss: 0.5335121370595077 (+/-) 0.03766979184623805 Precision: 0.6061427871581452 Recall: 0.7785517241379311 F1 score: 0.6816138984678044 Testing Time: 0.001272768809877593 (+/-) 0.0004460102877105528 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 7.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=7, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 7 No. of parameters : 175 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=7, out_features=2, bias=True) ) No. of inputs : 7 No. of output : 2 No. of parameters : 16 Dynamic laerning rate for each hidden layer: [0.02]
93% (28 of 30) |###################### | Elapsed Time: 0:00:00 ETA: 00:00:00C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.85517241379311 (+/-) 2.505247762115322 Testing Loss: 0.5316262460988144 (+/-) 0.03758203434551045 Precision: 0.6061427871581452 Recall: 0.7785517241379311 F1 score: 0.6816138984678044 Testing Time: 0.0013745735431539602 (+/-) 0.0006622602527701685 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
93% (28 of 30) |###################### | Elapsed Time: 0:00:00 ETA: 00:00:00C:\Users\SCSE\AppData\Local\Continuum\miniconda3\envs\stmicro\lib\site-packages\sklearn\metrics\_classification.py:1221: UndefinedMetricWarning: Precision is ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior. _warn_prf(average, modifier, msg_start, len(result))
=== Performance result === Accuracy: 77.85517241379311 (+/-) 2.505247762115322 Testing Loss: 0.5178975113506975 (+/-) 0.02604676863596477 Precision: 0.6061427871581452 Recall: 0.7785517241379311 F1 score: 0.6816138984678044 Testing Time: 0.0016168150408514615 (+/-) 0.000611088322322515 Training Time: 0.0 (+/-) 0.0 === Average network evolution === Total hidden node: 8.0 (+/-) 0.0 Number of layer: 1.0 (+/-) 0.0 === Final network structure === 1 -th layer hiddenLayerBasicNet( (linear): Linear(in_features=24, out_features=8, bias=True) (activation): Sigmoid() ) No. of inputs : 24 No. of nodes : 8 No. of parameters : 200 2 -th layer outputLayerBasicNet( (linearOutput): Linear(in_features=8, out_features=2, bias=True) ) No. of inputs : 8 No. of output : 2 No. of parameters : 18 Dynamic laerning rate for each hidden layer: [0.02]
========== Performance creditcarddefault ========== Preq Accuracy: 77.86 (+/-) 0.0 F1 score: 0.68 (+/-) 0.0 Precision: 0.61 (+/-) 0.0 Recall: 0.78 (+/-) 0.0 Training time: 0.0 (+/-) 0.0 Testing time: 0.0 (+/-) 0.0 ========== Network ========== Number of hidden layers: 1.0 (+/-) 0.0 Number of features: 7.8 (+/-) 1.94
In [ ]: